i
IN-SITU PASSIVE TREATMENT OF
MUNICIPAL SOLID WASTE (MSW) LEACHATE
USING A MODIFIED DRAINAGE LEACHATE
COLLECTION SYSTEM (LCS)
A Thesis Submitted to the College of
Graduate Studies and Research
in Partial Fulfillment of the Requirements
for the Degree of Master of Science
in the Department of Civil and Geological Engineering
University of Saskatchewan
Saskatoon, Saskatchewan, Canada.
By
Ernesto F. Ruiz
© Copyright Ernesto F. Ruiz, March 2005. All rights reserved.
ii
Permission to Use
In presenting this thesis in partial fulfillment of the requirements for a Postgraduate
degree form the University of Saskatchewan, I agree that the Libraries of this
University may make it freely available for inspection. I further agree that
permission for copying of this thesis in any manner, in whole or in part, for scholarly
purposes may be granted by the professor of professors who supervised my thesis
work or, in their absence, by the Head of the Department or the Dean of the College
in which my thesis work was done. It is understood that any copying or publication
or use of this thesis or part thereof for financial gain shall not be allowed without my
written permission. It is also understood that due recognition shall be given to me
and to the University of Saskatchewan in any scholarly use which may be made of
any material in my thesis.
Requests for permission to copy or to make other use of material in this thesis in
whole or part should be addressed to:
Head of the Department of Civil and Geological Engineering
University of Saskatchewan
57 Campus Drive
Saskatoon, Saskatchewan
Canada
S7N 5A9
iii
Abstract
This thesis describes a laboratory investigation of in-situ treatment of synthetic
leachate representative of that generated by a municipal solid waste (MSW) landfill.
The overall objective is to evaluate alternative designs and operating procedures for
effective leachate collection in conjunction with efforts to accelerate waste
stabilization (i.e. leachate recirculation). In the investigation five 15 cm (6”)
diameter PVC columns were packed with pea gravel and concrete of different sizes;
geotextiles were also placed between the packed sections as filter-separators and
promoters of bacterial growth. Synthetic leachate was continuously input to the top
of the columns and circulated at rates representative of operating field conditions.
For each column, effluent was discharged to a nitrification reactor before
recirculation. The tests were conducted under anaerobic and unsaturated conditions
in the columns. Results indicate about a 97% decrease in COD from the synthetic
leachate concentration entering the top of the column, and about 98 % conversion of
the ammonia to nitrogen gas. COD depletion and methane production were not
significantly inhibited by the denitrification process. Optimum Hydraulic Retention
Time (HRT) for the nitrification-denitrification system makes it economically viable
for its development at a landfill site. Gas production shows low CO2 values,
decreasing the potential of clogging in the Leachate Collection System (LCS) and
iv
extending the Landfill Gas (LFG) network’s life service by generating a less
corrosive environment. The use of concrete as an alternative to the most commonly
used natural gravel as leachate collection drains may not be a good option. During
the experiment, the leachate that permeated the columns packed with crushed
concrete, presented a higher pH than the leachate that permeated the natural stone.
At the conclusion of the experiment noticeable weathering was observed when the
columns where dismantled. Further studies are recommended until more conclusive
evidence as to concrete performance is found. The overall results obtained from the
experiment show that in situ passive treatment at landfills is viable.
v
Acknowledgements
I am in great debt to my parents, Hernán and Camila, for all their years of patience,
example, and support. They are the reason that this thesis was possible, and I know
they are proud of it.
I want to thank my supervisor Dr. Ian Fleming for his advice on real life and his
constant demand for practical engineering. I also want to thank my co-supervisor Dr.
Gordon Putz for his words of encouragement and patience at difficult times.
Moreover, I would like to thank them both for their financial support.
I am also grateful to Doug Fisher for his help and continuous exchange of ideas in
the environmental lab. Thanks to my fellow students Manoj Singh and Patrick
Schmidt for helping me with the laboratory tests and for being at the right place, at
the right time.
Lastly, to my wife Jean. Not only I am grateful for her constant support,
encouragement, and editing of my “spanglish” writing style, but also because her
wonderful way of looking at life makes it truly enjoyable.
vi
Table of Contents
Permission to Use ii
Abstract iii
Acknowledgements v
Table of Contents vi
List of Figures x
List of Tables xiii
Chapter 1 Introduction
1.1 Subject 1
1.2 Need 3
1.2.1 MSW Landfill Leachate Collection System (LCS)
in Bioreactor Landfills 4
1.2.2 Ammonia in a Bioreactor Landfill 5
1.3 Objectives 6
vii
Chapter 2 Background Information and Literature Review
2.1 Introduction 9
2.2 MSW Leachate Generation 10
2.3 MSW Leachate Characterization 11
2.3.1 Solid Waste degradation 11
2.3.2 Landfill mode of operation 13
2.3.3 Landfill design 15
2.4 MSW Leachate Management Strategies 17
2.5 MSW Leachate Treatment 19
2.6 Challenges of a Bioreactor landfill 22
2.7 MSW LCS as a Fixed Biofilm Reactor 23
2.8 Nitrifying and Denitrifying Bioreactors 26
2.9 Geotextiles in landfills 29
2.10 Conclusions 34
Chapter 3 Experimental Methods
3.1 Introduction 36
3.2 Column and Nitrification Reactor Set Up 36
3.3 Experiment Mode of Operation 41
3.3.1 Phase 1 41
viii
3.3.2 Phase 2 43
3.4 Physical Parameter Testing 43
3.4.1 Hydraulic Conductivity Testing 43
3.4.2 Porosity Testing 45
3.5 Water Quality Testing Program 56
3.6 Gas Production 58
3.7 Simulation of Leachate Generation and Recirculation 58
Chapter 4 Evaluation of Leachate Quality Using a Modified Drainage LCS
4.1 Introduction 60
4.2 Laboratory Testing Results 61
4.2.1 pH and Alkalinity 61
4.2.3 Ammonia Conversion to Nitrate 66
4.2.4 Nitrate Conversion to Nitrogen Gas 66
4.2.5 Total Nitrogen and Total Nitrogen Removal 72
4.2.6 COD Removal 72
4.2.7 Gas Production 79
4.3 Mass Balance Analysis 81
4.3.1 Nitrogen Mass Balance 81
4.3.2 Carbon Mass Balance 83
4.4 Conclusions 87
ix
Chapter 5 Hydraulic Performance of the Modified Drainage LCS
5.1 Introduction 90
5.2 Effects of leachate treatment in Geotextiles 91
5.2.2 Clog Material Analysis of Geotextiles 96
5.2.3 Calculating the Time for the Geotextiles to Clog 101
5.2.4 Consequences of Geotextile Clogging 105
5.2.5 Effects of Leachate Treatment on the Hydraulic
Performance of the Drains. 106
5.3 Potential of Recycled Concrete as a Drainage Material in a Landfill 109
5.3.1 Fines Accumulation 109
5.3.2 Leachate Mounding 111
5.3.3 Alteration of Leachate pH 112
5.3.4 Durability of Aggregate 113
5.4 Conclusions 116
Chapter 6 Discussion and Recommendations
6.1 Introduction 119
6.2 Discussion 119
6.3 Recommendations 123
References
x
List of Figures
Figure 2.1 Solid waste decomposition 13
Figure 2.2 Leachate collection system in a MSW landfill 30
Figure 3.1 Column schematic 37
Figure 3.3 Column discharge apparatus 40
Figure 3.4 Column C1 initial piezometric line 46
Figure 3.5 Column CC1 initial piezometric line 47
Figure 3.6 Column C2 initial piezometric line 48
Figure 3.7 Column CC2 initial piezometric line 49
Figure 3.8 Column C3 initial piezometric line 50
Figure3.9 Initial porosity profile of column C1 51
Figure 3.10 Initial porosity profile of column CC1 52
Figure 3.11 Initial porosity profile of column C2 53
Figure 3.12 Initial porosity profile of column CC2 54
Figure 3.13 Initial porosity profile of column C3 55
Figure 4.1 pH in anaerobic column discharge 62
Figure 4.2 pH in aerated effluent 63
Figure 4.3 Alkalinity in anaerobic column discharge 64
xi
Figure 4.4 Alkalinity in aerated effluent 65
Figure 4.5 Ammonia concentration in aerated effluent 67
Figure 4.6 NO3- concentration in the aerated effluent 68
Figure 4.7 Nitrogen species in aerated effluent (column C2) 70
Figure 4.8 Nitrogen species in anaerobic column discharge (column C2) 71
Figure 4.9 Total nitrogen in anaerobic column discharge 73
Figure 4.10Total nitrogen in aerated effluent 74
Figure 4.11 COD removal in anaerobic column discharge 76
Figure 4.12 COD removal in aerated effluent 77
Figure 4.13 Anaerobic column reactors gas production 80
Figure 5.1 Initial and final piezometric line of the columns 93
Figure 5.2 Uppermost geotextile in column C2 after 100 days of operation 94
Figure 5.3 Geotextile before the experiment 95
Figure 5.4 Geotextile after the experiment 95
Figure 5.5 Black slime on drains 106
Figure 5.6 Hardened gravel material 107
Figure 5.7 Cross sectional view of the gravel material 108
Figure 5.8 Geotextile clogged with fines originating from sieved, but not
washed, crushed concrete. 110
Figure 5.9 Graph representing leachate mounding above the top geotextile 111
Figure 5.10 pH history of leachate. (See chapter 4: pH in anaerobic column
discharge) 112
Figure 5.11 Slice of weathered stone with rock types labelled 113
Figure 5.12 Slice of weathered crushed concrete 114
xii
Figure 5.13 non-weathered concrete 115
Figure 5.14 weathered concrete 115
xiii
List of Tables
Table 2.1 Leachate treatment methods affordability 21
Table 3.1 Anaerobic column reactors content 38
Table 3.2 Geotextile characteristics 39
Table 3.3 Synthetic leachate main characteristics 42
Table 3.4 Composition of synthetic leachate 44
Table 3.5 Hydraulic conductivity of different materials in column C1 46
Table 3.6 Hydraulic conductivity of different materials in column CC1 47
Table 3.7 Hydraulic conductivity of different materials in column C2 48
Table 3.8 Hydraulic conductivity of different materials in column CC2 49
Table 3.9 Hydraulic conductivity of different materials in column C3 50
Table 3.10 Column C1 initial porosity 51
Table 3.11 Column CC1 initial porosity 52
Table 3.12 Column C2 initial porosity 53
Table 3.13 Column CC2 initial porosity 54
Table 3.14 Column C3 initial porosity 55
Table 3.15 Frequency of the water quality testing program 56
Table 4.1 Gas composition 79
Table 4.2 Nitrogen mass balance 82
Table 4.3 Carbon mass balance 84
xiv
Table 5.1 Permeability of the different sections of the columns C1, C2,
C3, CC1 and CC2 92
Table 5.2 Classification of clog material within the uppermost geotextile 97
Table 5.3 Clogging composition in the laboratory experiment vs. clogging
composition in the findings of Fleming et al. (1999) 99
Table 5.4 Slake test for non-weathered and weathered concrete 116
1
Chapter 1 Introduction
Safe and cost-effective management of Municipal Solid Waste (MSW) is a
significant environmental challenge for modern society. This chapter explains how
the challenge arises, what has been done in the past to deal with it, and how this
thesis approaches the issue. Furthermore, the main objectives of the thesis are
outlined, and a brief description of the content of each chapter is presented.
1.1 Subject
Despite intense government promotion of the 3R’s concept (Reduce, Reuse,
Recycle) as well as growing environmental awareness, in countries such as the
United States, still disposed more than 55% of generated solid waste in landfills
(USA EPA, 2002). Society continues to equate success with consumption and the
result of this ideology is wide scale waste production and disposal. Thus, there is a
growing urgency for environmental scientists and professionals to deal with the
problem associated with the reality that waste will, in large measure, continue to be
disposed of in landfill sites.
2
In Municipal Solid Waste (MSW) landfills, leachate is the contaminated liquor
generated as a consequence of the expulsion of liquid from the refuse. The cause of
this liquid expulsion can be compaction of the solid waste, degradation of organic
material, as well as, irrigation, precipitation, and/or percolation of water through a
landfill from sources such ground water or leachate recirculation (Quian et al. 2002).
Leachate contact with either surface or groundwater resources could contaminate
them. Leachate management is therefore an imperative to maintaining a healthy
environment.
Recent improvements in sanitary landfill design and operation have been focused on
barrier systems and the management of gas and/or liquid residues (Harris et al.
2000). Although improvements in lining systems have minimized the likelihood of
ground water contamination, they also have resulted in higher leachate recovery.
Leachate, which is typically high in organic content and dissolved solids, must be
treated before discharge in order to protect the environment. Leachate treatment has
a significant cost and has to be carried out during the active and post-closure period,
which may last for decades after the landfill has ceased accepting wastes. Among the
different schemes developed to deal with leachate treatment, those based on
traditional sanitary wastewater treatment are often applied in leachate management
(McBean et al. 1995). Several innovative and new applications of existing
technologies are now being continuously developed (Reinhart et al. 2002).
3
A novel approach in leachate management is the bioreactor landfill. In this approach,
refuse is not only contained but also rapidly decomposed. The method is based on
the principle of accelerated waste degradation through leachate recirculation and
wastewater anaerobic treatment. The bioreactor landfill was fist proposed by
Pohland in the early 80’s and has now emerged as an attractive solution for leachate
treatment. It has several attractive features including:
i. Rapid stabilization of the organic fraction in the waste;
ii. Maximization of landfill gas capture over a shorter time frame, thus
improving the economies of landfill gas (LFG) energy projects, and
consequently reducing green house gas emissions;
iii. Significant reduction in off-site transport of leachate for treatment or
disposal;
iv. Reduction of post-closure care and maintenance;
v. Increase of the volume of the landfill available for disposal as a result of
rapid settlement
The utilization of landfills as bioreactors also generates challenges. Increasing the
water content of the refuse by leachate recirculation can create engineering problems
such as landslides, surface seeps and/or liner failure (Qian et al. 2002). Accelerating
waste degradation can lead to clogging of the leachate collection system and
increasing ammonia concentration (Fleming et al 1999, 2002; Rowe et al. 2002).
4
1.2 Need
The need for this thesis is based on two assertions:
1. A bioreactor landfill offers potential benefits for MSW management.
2. Fleming et al. (1999), (2002) demonstrated that the Leachate Collection System
(LCS) of a landfill may play a central role in the composition and strength of the
leachate. Their investigation suggested that novel designs and operating conditions
for an efficient leachate collection and treatment, along with the understanding of
the chemical, physical, and biological phenomena that take place in a landfill LCS,
would greatly improve landfills operation and management.
The need for this thesis is therefore, to investigate the effects in leachate
composition that a bioreactor landfill can generate not only by leachate recirculation
but also by improving the design of LCS.
1.2.1 MSW Landfill Leachate Collection System (LCS) in Bioreactor Landfills
A landfill Leachate Collection System (LCS) is a drainage system that controls
leachate mounding (minimizing the risk of either spills or liner failure) and
transports the leachate to storage tanks for subsequent treatment. A LCS is
5
composed of a drain (normally local natural stone), a filter/separator element (in
most cases a geotextile) and a perforated pipe network.
In order to work as a bioreactor, the solid waste in a landfill has to reach field
capacity which can be understood as the water content at which any additional water
or liquid will drain out of the waste. Since many landfills in North America are
located in non-humid locations, the addition of external liquids (e.g. water,
wastewater, activated- sludge) is required to achieve field capacity. An increased
water content of the solid waste improves the ability of the microorganisms to
decompose organic material (Barlaz et al. 1989), and by this, increases the Chemical
Oxygen Demand (COD) depletion and the likelihood of clogging (Armstrong et al.
1998; Fleming et al.1999, 2004; Rowe et al. 2002; Paksy et al. 1998; Cooke et al.
1999, and VanGulck et al. 2003).
The constant flow of leachate through the drainage system generates physical,
chemical and biological processes. Particle encrustation (clogging) in the void space
of the LCS is the most adverse consequence of these processes (Fleming et al 1999,
2004; Armstrong, 1998; Cooke et al. 1999). However, Rittman et al (1996), (2003);
Fleming et al (1999), (2004); Rowe et al (2002) and VanGulck et al (2003), have
demonstrated that the production of clogging material is closely related (by product)
to the decrease in leachate organic matter concentration. Their laboratory
investigations of clog formation have shown COD depletion of up to 85% of its
original concentration in field and laboratory studies.
6
1.2.2 Ammonia in a Bioreactor Landfill
MSW has been estimated to contain about 4% protein (Barlaz et al. 1989) and
therefore, ammonia (NH3-N) is expected to be produced during the decomposition of
organic nitrogen. Since ammonia is stable under anaerobic conditions, it will
accumulate in leachate (Burton and Watson-Craik, 1998). Thus, high concentrations
of ammonia will persist long after the COD has decreased to concentrations
representative of well-decomposed refuse. Therefore, the treatment of leachate to
remove ammonia becomes an important aspect of long-term landfill management.
1.3 Objectives
The general objective of this study is to evaluate alternative designs and operating
procedures for effective leachate collection in conjunction with efforts to accelerate
waste stabilization (e.g. leachate recirculation). Laboratory columns simulating a
real landfill LCS were packed with different sizes and types of porous media and
leachate was recirculated at different rates representative of those found in a typical
North American landfill.
The specific objectives are as follows:
1. Investigate in-situ passive leachate treatment methods for the depletion of
COD, nitrogen and some inorganic compounds.
7
2. Analyse the hydraulic effects of in-situ passive treatment on the drains and
geotextile during the laboratory experiment.
3. Determine the feasibility of using concrete as a Leachate drain material.
In order to achieve these specific objectives, laboratory columns were operated as
fixed biofilm reactors. Their functions were twofold: to work as an organic
anaerobic digester and as a de-nitrifying reactor. Each one of the fixed biofilm
reactors was coupled with an external aerated nitrification reactor as an intermediate
oxidizing step in ammonia treatment and to provide further organic matter depletion.
The thesis is divided into six chapters. In Chapter 1, the objectives are introduced
along with a brief description of the topic. In Chapter 2, the literature review and
background information relevant to the investigation is presented. The nature and
origin of leachate are described, along with established approaches for leachate
characterization management and treatment. The principles of the bioreactor landfill
are introduced, highlighting the advantages and potential problems, as well as the
biological and geochemical aspects underlying the concept of the on-site passive
leachate treatment.
The laboratory test program is described in Chapter 3. The experiment set up is
shown and the methods for the physical and chemical analyses are described. In
Chapter 4, the experimental results of relations to water quality and gas production
are presented and analyzed. Chapter 5 is divided in two subchapters. In these
8
subchapters, the columns autopsies are described as well as the physical changes in
the different porous media used within the columns (natural gravel, concrete and
geotextile). Overall conclusions and recommendation for further studies are
provided in Chapter 6.
9
Chapter 2 Background Information and Literature Review
2.1 Introduction
Despite increasing environmental awareness, up to 55 % of the total Municipal Solid
Waste (MSW) generated in nations such as the United Sates is still disposed in
landfills (USA EPA 2002). The need for landfilling does not have a foreseeable end.
Other solid waste management alternatives such as incineration and recycling face
technical and social hurdles. Incineration is not a viable method of disposal for a
wide variety of wastes such as those with high moisture content (Harris et al. 2000).
Furthermore, incineration may lead to air pollution problems, and it creates an ash
residue that still must be landfilled (McBean et al. 1995). Recycling efforts
eventually encounter practical limits (since the concept relies heavily on voluntary
participation) that make further reductions in the waste stream hard to achieve (Qian
et al. 2002).
Among the existing solid waste management strategies, landfilling is the most cost-
effective (Kiely, 1997). A properly engineered landfill can deliver satisfactory
environmental and economic results. Municipal Solid Waste (MSW) leachate
management strategies are continuously being developed and/or improved. Among
them, on-site biological passive treatment is one of the most attractive leachate
10
management strategies due to its economic advantages (Reinhart et al. 2002;
Pholand and Kim, 1999).
This chapter outlines and explains the generation, characterization, management, and
treatment of leachate. It also describes recent investigations that support the
feasibility of passive treatment of leachate at the landfill site for removal of organic
carbon, nitrogen and certain inorganic species. The role of geotextiles in landfills as
filters and promoters of bacterial growth is also discussed.
2.2 MSW Leachate Generation
According to Farquhar (1989), leachate is created when moisture infiltrates the
refuse in the landfill, dissolving electrolytes, nutrients, and contaminants, and
producing moisture contents high enough to initiate liquid flow. The main source of
moisture is the percolation of water through the waste by irrigation, precipitation,
groundwater discharge, and/or leachate recirculation. Even if no water is allowed to
infiltrate into the refuse, a small volume of contaminated liquid forms due to
biological and chemical reactions (Quian et al. 2002).
11
Although it is unlikely that leachate will be generated at a constant rate throughout
the life of the landfill, it will follow a pattern similar to that of precipitation in the
region (Farquhar, 1989). Several water balance models such as the Water Balance
Method (WBM) and the HELP model (Hydrologic Evaluation of Landfill
Performance) provide insight into the trends of leachate production. However, due to
the difficulty of determining some of the coefficients in such models, uncertainty
may be associated with the result of such methodologies (McBean et al. 1995).
2.3 MSW Leachate Characterization
Most of the literature on leachate characterization claims that the main features of
leachate are determined by the degradation of solid waste over time and the landfill
mode of operation. This thesis argues that the design of the landfill and specifically
the LCS can also play a central role in the composition of leachate as it is actually
collected.
2.3.1 Solid Waste degradation
Approximately 80% of Municipal Solid Waste (MSW) is composed of anaerobically
biodegradable organic matter (Barlaz et al. 1989). The organic matter found in solid
12
wastes includes lignocellulosics (lignin, cellulose and hemicellulose), proteins, lipids
and starch. (Palmisano and Barlaz, 1996). Among them, cellulose and hemicellulose
are the major biodegradable constituents, comprising 45-60% of the MSW dry
weight (Barlaz et al. 2002). Thus, anaerobic MSW decomposition can be explained
largely by the transformation of cellulose and hemicellulose in simpler compounds
such as methane and carbon dioxide.
The transformation of cellulose and hemicellulose takes place in four different
phases (Barlaz, et al. 1989, Mcbean et al. 1995, Warith and Sharma, 1998). Phase I
or aerobic phase, involves a short period of aerobic decomposition in which easily
degradable organic matter is consumed and carbon dioxide is generated. During this
phase some polymer hydrolysis occurs, converting the initial cellulose,
hemicellulose and proteins in soluble sugar and aminoacids. This process is
mediated by extracellular enzymes secreted by microorganisms (Palmisano and
Barlaz, 1996).
In Phase two, the fermentation phase, sugars and proteins are fermented and
converted to volatile fatty acids and ammonia. This phase is characterized by the
presence of hydrogen producing and acetogenic bacteria as well as high levels of
carbon dioxide. Phase three is called the acid phase. Volatile fatty acids (long chain
carboxylic acids) are converted to short chain carboxylic acids (mainly acetate) by
hydrogen producing, hydrogen consuming and acetogenic bacteria (Parkin and
13
Owen, 1986). Phase four is the methanogenic phase. Here, methanogenic bacteria
convert the hydrogen, acetate and some of the carbon dioxide generated during the
previous phases into methane gas. Figure 2.1 summarizes the decomposition phases
described above.
Complex Organic Solids• Carbohydrates
• Proteins
• lipids
Simpler Soluble Organics
1
Propionate,
Butyrate, etc.(Long chain
fatty acids)
1
H2 – CO2 Acetate
CH4 – CO2
11
2 2
3
4 5
Bacterial Groups:1. Fermentative
2. Hydrogen-producing,Acetogenic
3. Hydrogen-consuming, Acetogenic
4. CO2-Reducing methanogens
5. Aceticlastic methanogens
Figure 2.1 Solid waste decomposition (after Parkin and Owen, 1986)
2.3.2 Landfill mode of operation
Christensen et al. (1992) have described several individual management procedures
and their effects on waste degradation (which is reflected in methane production):
14
Waste Composition: The composition of the waste will determine the rate at which
the waste decomposes. For instance, an increased content of newspaper does not
affect gas generation significantly, while an increased content of magazines increase
the gas production (Stegmann and Spendlin, 1986). Large concentrations of
sulphate increases the redox potential, and sulphate reduction may compete with
methanogenic bacteria for organic carbon, thus decreasing methane gas production
(Christensen and Kjeldsen, 1989).
Sewage Sludge: Positive results in methane generation due to the addition of sludge
to the refuse, depends on the type of sludge added. A sludge with low pH (e.g. septic
sludge) may have a negative effect in methane formation (Leckie et al. 1979), while
neutral well-buffered sewage sludge may have positive effects (Leushner, 1989).
Buffer Addition: Despite some beneficial effects of buffering in waste degradation
(Stegmann and Spendlin, 1989), it does not seem necessary in all landfill situations.
However, if a landfill has failed to generate methane due to low pH values, the
addition of a buffer is an obvious measure to help establish methanogenic conditions
(Christensen et al.1992).
Shredding: Smaller particle size can increase the rate of the hydrolysis of the
organic wastes as a result, the acid phase is intensified resulting in increasing
production of carbon dioxide, low pH, and high content of organic carbon in the
15
leachate with the consequences of no or reduced methane formation (Barlaz et al.
1989).
Compaction: Compaction of the refuse is necessary due to the need for optimum
use of the landfill capacity and for obtaining geotechnical stability. Results obtained
in filed and laboratory experiments may indicate that the time of compaction of the
upper refuse layer may be seen as a possible control of the acid phase and hence it
can delay the initiation of the methanogenic degradation (Christensen et al. 1992).
Recirculation of Leachate: This concept consists of removing the leachate from the
base of the landfill and then to reintroduce it onto, or into, the waste mass by any
one of a number of methods, such as: surface spraying, surface ponding, leach fields,
shallow wells and/or deep wells (Quian et al. 2002). Pohland (1980) noted that the
daily recirculation of the leachate provides microorganisms with sufficient nutrients
and, as a result, overall conversion of the waste is enhanced.
2.3.3 Landfill design
Several researchers have found that leachate may undergo significant treatment after
passing through a porous drainage medium. Fleming at al. (1999) demonstrated this
phenomenon with forensic excavations and leachate analysis from the Keele Valley
Landfill (KVL) site near Toronto (Canada). Peeling et al. (1998); Paksy et al.
(1998); Fleming et al. (1999); Fleming and Rowe (2004); and VanGulck et al.
16
(2003) supported those findings using laboratory LCS drains simulating real landfill
conditions.
Due to practical difficulties extracting the leachate directly from the base of the
buried waste, leachate is taken from the end of the leachate collection pipes. Before
reaching the end of the leachate collection pipes, during what may be a residence
time of months or years (Fleming et al. 1999), biological stabilization of the leachate
by biofilm (slime) will significantly change the leachate composition. Fleming et al
(2002) hypothesised that the leachate collected at the end of the LCS system of
landfills has undergone a degradation process within the “anaerobic-fixed biofilm
reactor” represented by the slime-covered drainage blanket underlying the waste.
To support that statement, they compared two leachate samples coming from two
landfills with similar characteristic of age and type of refuse in Ontario (Canada).
One landfill has a leachate collection system (LCS), the other one does not. The
landfill with a LCS presented an ongoing depletion of Chemical Oxygen Demand
(COD) when compared to a stable contaminant such as chloride, whereas in the
landfill without a LCS, COD and chloride appear to both increase and decrease in
parallel.
It can be hypothesized then, that the leachate coming from the landfill with no LCS
underwent waste decomposition dominated by the natural sequence of solid waste
degradation phases as described in section2.3.1. In comparison, the landfill with a
17
LCS provided a more steady treatment of organic matter in the leachate due to the
additional degree of degradation which occurred in the biofilm-rich LCS. The
difference is important especially when trying to achieve rapid waste stabilization
may bring substantial economic benefits and environmental regulatory compliance
for the landfill’s owner/operator.
Waste decomposition, and thus the leachate composition, is governed by the basic
processes and the solid waste degradation phases described in section 2.3.1, but the
LCS of landfills can play an important complementary enhancing role in dealing
with leachate (due to the possibility of working as a fixed biofilm reactor) by
creating a more controlled and efficient in-situ passive leachate treatment system.
2.4 MSW Leachate Management Strategies
Leachate management strategies can be classified as: natural attenuation,
remove/treat/discharge, or leachate recycling (Quian et al. 2002).
Natural Attenuation: This was the most commonly adopted strategy for leachate
management until the early eighties when regulatory agencies issued more stringent
rules for leachate handling. This was due to uncertainties regarding the hydraulic
conductivity and adsorption capability of the soil acting as a barrier (mostly clay
soils). Today, there seems to be a resurgence in the utilization of natural attenuation
in arid climates where the amount of leachate produced is low. Nonetheless, some
18
factors remain difficult to manage under this strategy. Among the most important
are: leachate strength and composition, and landfill site selection (it must be located
in an isolated area with the consequence of increased hauling costs) (Barlaz et al.
2002).
Remove/Treat/Discharge: This is the most common strategy for liquid
management in current landfills. The principle is to avoid any contact of the leachate
with the surrounding environment. In order to do this, the waste is “encapsulated” by
impermeable barriers which can be natural, synthetic or a combination of the two.
Leachate is removed from the base of the landfill and conveyed to containment
facility for later treatment and disposal. This approach of the containment system
requires for 30 to more than 100 years post-closure.
Leachate Recycling: This concept is based on the finding that solid waste
degradation is a dynamic process, primarily influenced by waste characteristics,
availability of moisture and nutrients, and current operational circumstances. The
last three can be assisted by controlled accumulation, containment, collection and
recirculation of leachate back through the landfill. This management strategy offers
opportunities to establish the necessary mutually advantageous relationship between
the microbial population acting during the acid and methane production phases of
landfill stabilization, thus enhancing reaction rates and conversion pathways in a
shorter and more predictable manner (Pohland and al-Yousfi 1994).
19
A very attractive feature of leachate recycling from the economic perspective of
owners and/or operators is that this will allow them to swiftly achieve
“sustainability” which occurs when they are completely able to manage the outputs
(liquids and gases), provide environmentally acceptable residues, avoid long post-
closure care periods, and can potentially use the closed sites for beneficial purposes
(Reinhart et al. 2002).
2.5 MSW Leachate Treatment
According to Kiely (1997), the selection of leachate treatment and disposal
alternatives depends upon:
1. The estimated leachate generation rates.
2. Physical-chemical features of the leachate and variations in leachate
characteristics/flow over time.
3. Identification of final disposal alternatives and evaluation of their feasibility
in terms of cost effectiveness, environmental impact, technological
constraints, regulatory requirements, and compatibility with other elements
of the landfill design and operation.
4. Estimated capital, and operation and maintenance costs of treatment and
disposal method with respect to reliability and flexibility, and
5. Age of landfill.
20
The most common alternative for leachate treatment is the off-site treatment at a
Municipal Waste Water Treatment Plant (MWWTP) (Harris et al. 2000). Although
this solution is typically economical, problems with biological treatment may be
experienced at the MWWTP. Another disadvantage is the cost of hauling leachate.
There are also innovative technologies which are normally borrowed from different
and bigger industries, such as food and petrochemical. Among them are reverse
osmosis, thermal oxidation, air stripping. The use of such approaches is, in many
cases, not affordable by the solid waste market (Harris et al. 2000).
On-site passive treatment, on which this thesis is focused, relies primarily on
biological systems. This type of leachate treatment is becoming increasingly
attractive due to lower costs of operation and maintenance. Among the on-site
passive treatment options, the bioreactor landfill mode of operation is arising as one
of the most investigated and promoted in North America by the landfill industry
(Quian et al.2002). Although, a promising technology, it still has to overcome
several challenges that arise due to operational issues, such as clogging of the LCS
(Fleming et al. 1999), increased ammonia concentration in the leachate (Pohland,
1995) and the possibility of problems with the landfill stability and slope failure
(Quian et al. 2002).
Table 2.1 illustrates the main features of different leachate treatment methods and
their affordability compared to a sequencing batch reactor system which is a system
21
Table 2.1 Leachate treatment methods affordability (After Harris et al. 2000).
System Description Major Unit Process Total Relative Cost* Conventional Equalization
pH adjust-Chemical Precip-
Sedimentation
Biological, SBR
Residual Management
5.0
Conventional Equalization
pH adjust-Chemical Precip-
Sedimentation
Biological, Fixed Film
(Packed Towers, Trickling
Filters, RBC)
Residual Management
7-7.5
Conventional Lagoon
Residual Management
1.75-2.25
Membrane Equalization
pH Adjustment
Pre-Filtration
Reverse Osmosis
Residual Management
4.75-5.55
Thermal Equalization
Evaporation
Thermal Oxidation
Residual Management
3.3-3.9
Thermal Equalization
pH Adjustment
Distillation
Residual Management
4.0-4.75
Biological –In Situ
(Bioreactor)
Equalization
Recirculation-Moisture
Content Control
LF Gas Control
2.25-2.75
Biological –In Situ
(Facultative Bioreactor)
Equalization
Nitrification
Recirculation-Moisture
Content Control
LF Gas Control
2.9-3.5
Biological-Land Base Equalization
pH Adjustment
Constructed Wetlands
2.0-2.6
Biological-Land Base Equalization
pH Adjustment
Phytoremediation
4.4-5.0
*Scale 1-8, SBR (Sequencing Batch reactor) = 5, with 1 = easiest to operate and
lowest capital cost and O&M.
22
that has been successfully used to treat municipal and industrial waste water as well
as MSW leachate (US EPA, 1999)
It can be seen in table 2.1 that the biological in-situ based treatments are among the
least expensive systems, which is expected since the treatment of the leachate take
place on- site and is carried out by naturally ubiquitous bacteria.
2.6 Challenges of a Bioreactor landfill
The rapid stabilization of the waste in a landfill by leachate recirculation brings not
only benefits but also challenges. Increasing ammonia levels and clogging of the
LCS are two of the most important issues for a landfill that operates as a bioreactor.
Rittman et al. (1996),(2003), Rowe et al. (1997), (2000), (2000a), (2000b), (2002),
and VanGulck (2003), have found a direct relationship between anaerobic organic
stabilization of the organic load (expressed as COD) and the precipitation of calcium
carbonate and other minerals in the LCS. Furthermore, MSW has been estimated to
contain about 4% protein (Barlaz et al. 1989), therefore, ammonia (NH3-N) is
expected to be produced during the decomposition of organic nitrogen and since
ammonia is stable under anaerobic conditions, it will accumulate in leachate (Burton
and Watson-Craik, 1998).
Similarly, in a landfill with leachate recirculation, with out any mechanism for
nitrogen removal, ammonia concentration in the leachate will increase until the point
23
at which it may interfere with the effectiveness of ongoing acetotrophic COD
removal (Fleming et al. 2002).
Tchobanoglous et al. (1993) presents a sequence of reactions for solid waste
decomposition that may shed additional light in understanding from where ammonia
and carbonates come:
Solid Waste + H2O + Nutrients → New cells + Resistant organic matter + CO2 +
CH4 + NH3+ H2S + heat (2.1)
If acceleration of waste degradation is to be achieved through leachate recirculation,
it becomes apparent that both issues (the clogging of the LCS and the accumulation
of ammonia) must be addressed in order to achieve satisfactory leachate
management results.
2.7 MSW LCS as a fixed biofilm reactor
The discovery of solid particles (mainly composed of calcite and iron sulphide)
entrapped in a LCS in Germany by Brune (1991) prompted several researchers to
investigate the phenomenon due to the potential implications for the long term
serviceability of landfills. The presence of precipitates in the void spaces of the drain
and /or the obstruction of the leachate collection pipes has the potential to create
leachate mound, with the consequences of additional stress on the containing barrier
24
(increased advection) and decreased landfill slope stability as a result of increased
pore pressures.
Fleming et al. (1999) observed leachate monitoring, field excavation of the landfill
drainage system, and analysis of the clogging material at the Keele Valley landfill
site (KVL) (Ontario, Canada). It was found that a thick black slime layer built up on
the 50 mm diameter clear stone drainage blanket underlying the KVL after 1 to 4
years of exposure to leachate. Its amount was visibly higher near the perforated
drainage pipes where there is higher flow rate of leachate. Also within the pipes,
significant precipitation of deposits resulted in the accumulation of large solid
mineral clog structures (up to 30 cm diameter).
It was hypothesised that the physical, chemical and biological processes occurring
during the anaerobic organic degradation of the waste have a strong relationship
with biofilm and clog formation. Further laboratory studies simulating MSW
leachate collection drains (Peeling et al. 1998; Paksy et al. 1998; Fleming et al.
1999; Fleming and Rowe 2004; Armstrong 1998 and Rowe et al. 2002) and
theoretical models (Cooke et al. 2001 and Rittman et al. 2003) confirmed that
hypothesis. The organic waste in a landfill will be reduced to simpler organic
compounds such as carboxylic acids (mainly acetic acid). Methanogenic bacteria
will consume those acids, oxidizing them into a carbonate form (mainly CO2).
Approximately 50% of the CO2 will off-gas from the leachate, and the other 50 %
will be hydrolyzed back into the leachate in the major form of CO2(g) or H2CO3*
25
(Rittman et al. 1996). The change of a weak acid (acetic acid) for an even weaker
one (H2CO3*) will increase the pH, shift the total carbonate composition from
H2CO3* to CO32- and saturate (even supersaturate) the leachate with carbonate,
which precipitates as calcium carbonate (CaCO3). Much of the bacterial population
that brings about the anaerobic leachate degradation, develops within the drain
(gravel or washed quarry stone), where the biofilm grows on the surfaces and is in
contact with the organic-rich leachate. The precipitated calcium carbonate, along
with other minerals such as iron sulphide and residues of fine material from the
crushed stone and/or daily cover soil (Bennet et al. 2000), will then be entrapped
into the biofilm forming what is called a clog.
Based on the previous findings, it can be said that calcite precipitation in a LCS is
intimately associated with organic matter stabilization (i.e. COD removal from the
leachate). Thus, a beneficial process is associated with a significant operational
challenge. The positive event of organic matter degradation (and in the case of
leachate recirculation, an accelerated one), could lead to a negative side effect such
as the obstruction of the LCS of the landfill. Fleming (1999); Fleming and Rowe
(2004); Cooke et al. (2001), Rowe et al. (2002); Rittman et al. (2003) and VanGulck
(2003), quantified the connection between organic matter stabilization and clog
formation. This connection was called the Yield Coefficient (Yc) and relates the
amount of calcium (or carbonates) precipitated with the depletion of Chemical
Oxygen Demand (COD) which is a general indicator of the organic matter
concentration in the leachate.
26
2.8 Nitrifying and Denitrifying Bioreactors
Increased levels of nitrogen in the form of ammonia as a result of acceleration of the
waste degradation is expected when leachate is recirculated (Knox, 1985). The
major concerns of untreated nitrogen discharges include eutrophication of receiving
waters, toxicity to aquatic life, dissolved oxygen depletion in receiving waters, and
possible contamination of ground water. Fleming et al. (2002) also suggested that
ammonia can affect the efficiency of the LCS to work as a fixed biofilm reactor.
Thereby, ammonia treatment becomes an imperative.
Ammonia naturally degrades by two well known phenomena: nitrification and
denitrification. The former takes place in the presence of oxygen (aerobic) and the
latter in its absence (anaerobic). Since a landfill can provide both environments, it
seems plausible to attempt to remove ammonia at the landfill site. A LCS has very
low levels of oxygen (it has been depleted by bacteria) and might work as a
denitrification reactor. On the other hand, an external concrete tank for example,
could also work as a nitrification reactor.
In order to determine the feasibility of in-situ ammonia treatment, it is important to
understand what nitrification and denitrification are and how they can be
implemented.
27
Nitrification is an autotrophic aerobic process which utilizes an inorganic carbon
source (carbonates), an inorganic electron donor or energy source (NH4+ or NO2
-),
and elemental oxygen as a terminal electron acceptor. The complete oxidation of
ammonium to nitrate occurs in two intermediary steps by two different genera of
autotrophic bacteria (US EPA 1975):
Nitrosomonas:
55 NH4+ + 76O2 + 109HCO3
- → C5H7NO2
- + 54NO2
- + 57H2O + 104H2CO3
(bacteria)
(2.2)
Nitrobacter:
400 NO2- + NH4
+ + 4H2CO3 + HCO3
- + 195O2 → C5H7NO2
- + 3H2O + 400NO3
-
(bacteria)
(2.3)
Heterotrophic denitrification is an anoxic process which utilizes an organic carbon
source (such as methanol) for synthesis and as an electron donor, and nitrite or
nitrate as the terminal electron acceptor. (Azevedo et al. 1995). Complete
denitrification occurs in two steps by a broad range of facultative bacteria such as
Pseudomonas, Micrococcus, Archromobacter, and Bacillus. Equations for nitrite and
nitrate reduction can be represented as follows (US EPA 1975):
28
Nitrate reduction
NO3- + 0.33CH3OH → NO2
- + 0.33H2O + 0.33H2CO3 (2.4)
(methanol)
Nitrite reduction
NO2- + 0.5CH3OH + 0.5H2CO3 → 0.5N2 + HCO3
- + H2O (2.5)
(methanol)
Biological treatment of leachate with high ammonia concentration has been
investigated by a number of researchers. Azevedo et al. (1995) found that a
biological reactor (Modified Ludzack Ettinger-MLE) can treat ammonia levels of up
to 1500 mg/L and that temperature plays an important role. Nitrification and /or
denitrification ceased below a temperature of 10˚C due to inhibition of bacterial
activity.
Shiskowsky and Mavinic (1998) and Price et al. (2003), showed the importance of a
source of carbon for heterotrophic denitrification. Onay and Pohland (1998)
investigated nitrogen management in bioreactor landfills concluding that leachate
recirculation increased uniformity of moisture, substrate, and nutrient distribution
creating an environment that promoted the rapid development of the desired
microbial population of denitrifiers, nitrifiers and methanogens.
The denitrification potential of actively decomposing and well decomposed refuse
was measured by Price et al. (2003). Results showed that nitrate did inhibit methane
29
production (microorganism obtain more energy for growth under nitrate-reducing
conditions relative to methane-producing conditions. ∆G˚= - 1120.5 kJ and -31 kJ,
respectively). However, the reactors recovered their methane–producing activity
with the termination of nitrate addition. It was also hypothesized that nitrate,
because of the large scale of a landfill, and the tendency of the different processes to
dominate in different zones of the landfill, will not likely have a significant effect on
methane yields in full-scale landfills. Therefore, landfills have significant capacity to
convert nitrate to nitrogen, which can be safely released to the atmosphere.
2.9 Geotextiles in landfills
A geotextile can be defined as any permeable textile used in a geotechnical
engineering work as an integral part of a man made project, structure, or system.
Geotextiles are made from synthetics fibres such as polyester, polyethylene, or
polypropylene and are usually classified as knitted, woven, or non-woven products
(Polarczyk 2000).
The main usage of geotextiles in landfills is as a filter over various drainage
materials such as soil materials, geonets, leachate collection pipes, leachate
collection trenches (Quian et al. 2002). The non-woven needled punched
30
geotextiles are the most commonly used in landfills as a filtration layer (Koerner and
Koerner, 1995). Geotextiles are also use in landfills to separate the different material
that comprise a LCS. Since filtration is one the most important uses of geotextiles in
landfills, clogging of the geotextile appears as a very important factor of
consideration. Figure 2.4 shows a MSW LCS typical design and the placement of
the geotextile.
Figure 2.2 Leachate collection system in a MSW landfill. (After Craven et al. 1999)
Reinhart and Chopra (2000) described three types of clogging that can affect LCS of
landfills and, therefore, geotextiles.
31
Chemical clogging
External phenomena such as the access of air and a drop in temperature can lead to
changes in the solubility of compounds in leachate. For instance, leachate containing
iron in contact with oxygen will oxidize the soluble Fe+2 into the insoluble Fe
+3.
Also, a decrease in temperature may affect the solubility of certain salts (Ramke,
1989).
Another possible source of chemical clogging comes from changes in leachate
chemistry which may lead to the saturation of carbonate in leachate and its
precipitation in the form of calcite or hydroxides compounds (Rittman et al. 1996;
Fleming et al. 1999, Rowe et al. 2002 and VanGulck 2003; Koerner et al. 1998,
Halse et al. 1987 (part 1 and 2).
Particulate Clogging
According to Reddi (1997) particulate clogging can be divided in two main
categories and both can be present in a landfill. The first one is a straining
mechanism which occurs when the size of the filtration media is similar to that of
the suspended particles resulting in a cake formation on the media.
The second one is the non straining mechanism where the driving forces of transport
and removal between the particles and the media are physicochemical. This non
32
straining mechanism can be of three types: Interception, which is the result of the
collision of suspended particles with the fabric of the filtration media; Sedimentation
which is the result of density differences between the suspended particles and water,
and Brownian motion, which refers to the movement of micron and sub-micron
particles due to diffusivity (Reinhart and Chopra, 2000)
Biological Clogging
Microbial biofilm consist of cells entrapped within a gelatinous matrix of extra
cellular polymers (EP) produced directly from the surface associated
microorganisms. It is believed that cells consume nutrients and redirect a portion of
the substrate to exopolymer production, binding cells to the surface of the fibre. The
EP acts as a cementing agent to reinforce cell binding to a surface (Polarczyk 2000).
Needle –punched non-woven geotextiles have been used successfully as bacterial
support in aerobic and anaerobic up-flow reactors to treat domestic and industrial
waste effluents. Because of their large porosity, their surface characteristic, the
specific area of their fibres, the size of their pores and their permeability, these
geotextiles offer many advantages for the development of bacterial biofilm growth
(Rollin and Lombard 1988).
Based on the tendency of geotextiles to develop a bacterial biofilm growth,
geotextiles are prone to clogging when permeated with a liquid with high organic
33
matter content such as leachate. Knowing this, Koerner and Koerner (1992) and
Koerner (1994) investigated several geotextile filters and soil filters, with different
leachate strength, and compared their responses to water as a permeant. Their
general finding is that geotextiles permeated by leachate with more than 2,500 mg/L
of TSS or BOD required laboratory simulation to assess the severity of the clogging
before placing them in the landfill.
Another source of biological clogging is through the degradation of the geotextile
due to attachment and attack by microorganisms such as mould, mildew and fungi.
They attack some types of finishes applied to textile fibres without attacking the
fibres themselves. Among the materials used in geotextile manufacture, polyesters
and polyolefins provide good biological resistance, while polyamides are known to
be mildly attacked by mildew and bacteria (Gallagher, 1998).
Biological growth by itself, as discussed above, certainly creates a decrease in the
ability of geotextiles to be permeated but, it is the mineral solid entrapment in the
biofilm created by bacteria that aggravate the circulation problem. As Mlynarek and
Vermeersch (1999) clearly explain:
“Solid retention phenomenon is an additional factor in biological
activities. At the filter interface, the solids, together with the
microorganisms, are part of a dynamic system where all components
interact with each other. This biofilm can combine with other biofilms
around agglomerated particles. In well-aerated systems, aerobic
organisms are active in breaking down the organic compounds while
34
anaerobic conditions, such as those found in a LCS, favour
encrustation. The result is the formation of a biofilm that may proceed
to impede flow so much as to clog the system”.
2.10 Conclusions
Significant improvements have been made in leachate management during the past
20 years. The natural development in landfill research seems to be on issues related
to post-closure time and minimisation of long term liabilities. Through proper
operating conditions and innovative designs, bioreactors can act efficiently and
shorten the time that waste needs to decompose. The advantages of reaching quicker
stabilization are numerous including: maximization of landfill gas capture for energy
projects, lower cost of leachate treatment and disposal, landfill space capacity reuse
as a result of rapid settlement, and a reduction in landfill post-closure care and
maintenance.
Nitrification, denitrification and organic matter degradation are phenomena well
understood. They can play a central role in the objectives of reaching faster
stabilization in landfills. The laboratory experiment described in the following
chapters, demonstrates that these phenomena can be fully developed in the leachate
collection system of a landfill and using an external aerated nitrification tank of an
economically and technically feasible scale.
35
With this in mind, this thesis attempts to find a new role for LCS in landfills. Drains
and geotextiles can be utilized as an engineered bacterial growth promoter with
beneficial decreases in the resulting leachate strength. A combination of an effective
drain-filter system with high surface area could play two important roles: to treat
leachate while avoiding or mitigating the negative consequences of excessive
clogging.
36
Chapter 3 Experimental Methods
3.1 Introduction
This chapter describes the experimental set up and the laboratory testing program. It
presents the rationale for the different designs and operating conditions, a
description of the initial hydraulic properties, and the methods used for the testing
programs.
3.2 Column and Nitrification Reactor Set Up
The laboratory testing program described in this thesis used five 15cm (6”) diameter
columns labelled C1, CC1, C2, CC2, C3 constructed from modular sections. Each
column was made of five PVC bolted flanged "spool" sections 40 cm high. (See
Figures 3.1). The uppermost “spool” section was used to contain a 15 cm layer of
refuse. The refuse was dug up from an old closed landfill in Ontario (Canada). The
main function of the refuse, was to inoculate the column with microorganisms and
particulate matter since they are absent in the synthetic leachate (See Table 3.3). The
material was cut, uniformly mixed, and placed on top of the uppermost geotextile in
each column.
37
Since the size of the drain material has been shown to be an important controlling factor
in the process of clog formation within the LCS (Armstrong 1998, Rowe et al, 2000), the
lower four sections of the columns were each packed with a different size and type of
granular drainage media as listed in Table 3.1.
Figure 3.1 Column schematic
A large surface area increases the ability of microorganisms to attach and to grow, and
increases the efficiency of the “passive treatment” COD removal (Fleming and Rowe,
38
2004). However, the likelihood of clog formation is greater for finer and more well
graded material. Thus, each column represents a multilayer design in which each layer of
finer material (for more treatment) are underlain by coarser layers (less susceptible to
clogging).
Table 3.1 Anaerobic column reactors content
Section Column C1 Column CC1 Column C2 Column CC2 Column C3
S1 Refuse Refuse Refuse Refuse Refuse
F1 Geotextile
(1200R)
Geotextile
(1200R)
Geotextile
(1200R)
Geotextile
(1200R)
Geotextile
(1200R)
S2 37-50 mm
gravel
20-25 mm
WSC concrete
37-50 mm
gravel
20-25 mm
WSC concrete
37-50 mm
gravel
F2 Geotextile
(1200R)
Geotextile
(1200R)
Geotextile
(1200R)
Geotextile
(1200R)
Geotextile
(1200R)
S3 16-19 mm
gravel
16-19 mm
gravel
9-12.5 mm
gravel
9-12.5 mm
gravel
9-12.5 mm
gravel
S4 9-12.5 mm
gravel
9-12.5 mm
gravel
5-9 mm
gravel
5-9 mm
gravel
5-9 mm
gravel
F3 Geotextile
(1200R)
Geotextile
(1200R)
Geotextile
(1200R)
Geotextile
(1200R)
Geotextile
(1200R)
S5 37-50 mm
gravel
20-25 mm
WSC concrete
37-50 mm
gravel
20-25 mm
WSC concrete
37-50 mm
gravel
WSC concrete = washed-screen-crushed concrete
Recycled WSC concrete was also used in two columns (CC1 and CC2) in order to
determine its suitability as a replacement for natural gravel as the drainage media. This
was investigated since some landfills may have the possibility of using recycled concrete
as a cheaper alternative.
The concrete was supplied by the City of Regina (Canada). The nominal diameter of the
crushed concrete was somewhat smaller than that of the natural stone used in the top
39
drainage layer. The concrete material contained fines in significant quantity which made
it necessary to wash it thoroughly in order to avoid a rapid plugging of the geotextiles.
Geotextiles (Terrafix 1200R TM) were placed as indicated in Table 3.1. Their main
functions were to work as filter, separator and promoter of bacterial growth. Koerner and
Koerner (1992) demonstrated that geotextiles are suitable environments for the growth of
microorganisms. Its main features are listed in Table 3.2
Table 3.2 Geotextile characteristics
Property MARV (minimum average roll
value)
GrabTensile
Newtons (N) CAN/CGSB-148.1
No 7.3-92
1,550
Mullen Burst –
Megapascals (Mpa)
CAN/CGSB-4.2
No.11.1-94
3.80
Tear Propagation
Newtons (N) CAN/CGSB-4.2
No.12.2-95
700
F.O.S. (um)
Micrometres
50 to 150
Permeability K (cm/sec)
CAN/148.1
No. 14-94
1.5 X 10-1
Elongation %
at Break
45 to 105
Standard Roll
Sizes Metres (m)
3.5 X 50
Standard Roll
Weight (lbs)
285
Manometers were installed along the length of the columns to monitor piezometric head
and to allow for leachate sampling, if desired. TedlarTM bags with a volume capacity of
1L were placed at the top of every column for gas sampling.
40
Each column was also equipped with an external nitrification reactor that received
leachate collected from the bottom of the column and which was completely aerated
before it is recirculated to the top of the column. (See Figure 3.1).
Plastic 4 L pails were used as nitrification reactors vessels. Initially the column discharge
was directly connected to the nitrification reactor. This initial system did not work out as
expected because leachate was released to the nitrification reactor in batches and not
continuously, resulting in upsets to the nitrification process. Therefore, 250 ml
Erlenmeyers were used as column discharge collectors and hydraulic seals, allowing the
column leachate discharge to be kept anaerobic and to be delivered steadily into the
aerated nitrification reactor. (See figure 3.3)
Figure 3.3 Column discharge apparatus.
Column
discharge
To nitrification
reactor
Column
41
The columns and the aerated nitrification reactors were operated in an environmental
control chamber maintained at a constant temperature of 35°C which is within the
temperature range that can be found during the anaerobic decomposition of solid waste at
the LCS (Barone et al. 2000).
3.3 Experiment Mode of Operation
The experiment was divided into two phases. An initial phase 1 or preparation phase and
a final phase 2 or the actual experiment.
3.3.1 Phase 1
The objective of the start-up phase 1 of the experiment was to inoculate bacteria inside
the columns. In order to do so, a mixture was prepared with a microbial consortium
representative of landfill conditions.
The mixture consisted of synthetic leachate mixed with leachate-saturated auger cuttings
collected from boreholes drilled at a closed landfill (Brock west landfill, Ontario,
Canada). The composition of the synthetic leachate was based on the work of Rowe et al
(2002). The main characteristics of the synthetic leachate are given in Table 3.3. The
mixture was incubated at 35°C for 3 days, then filtered with a geotextile (Terrafix 1200R)
and circulated through the columns at a rate of 1 to 2 L/day for 120 days under
unsaturated conditions. After this initial inoculation treatment the COD removal from the
42
leachate was approximately 60%, and BARTsTM (Biological Activity Reaction Test)
showed that a stable and large bacterial population (approximately 5,000,000 cfu/mL)
was present.
Concurrently, nitrifying bacteria were grown in the laboratory for use as an inoculating
culture for each column’s external nitrification reactor. In order to start up these
nitrification reactors, sludge coming from the City of Saskatoon waste-waster treatment
plant was used as a source of nitrifying bacteria.
Table 3.3 Synthetic leachate main characteristics
Characteristic Content or Level
COD 17,345 mg/L
Hardness 4,400 mg/L as CaCO3
Alkalinity 5,430 mg/L as CaCO3
TDS 15,100 mg/L
Eh -150 mV
pH 6
Total Nitrogen 766 mg/L as N
Calcium 2,900 mg/L as CaCO3
Chloride 3,791 mg/L
Sodium 4,429 mg/L
Volatile Fatty Acids 13,000 mg/L
TOC 5,800 mg/L as C
Approximately 3 weeks after the sludge was introduced into the aerated vessel, and the
reactor was fed with ammonia, nitrifying bacteria were detected indirectly through the
presence of nitrite, and later nitrate, which confirmed the complete oxidation of ammonia.
43
3.3.2 Phase 2
Once the column inoculation and start-up phase was completed, synthetic leachate (see
Table 3.4) which has been kept under anoxic conditions, and at a temperature of 4oC,
was pumped to the top of each column at a rate of 100 mL/day. The concentration of the
nitrification vessel was diluted from 3L to 15 L with distilled water and then evenly split
amongst the five external 3 L nitrification reactors receiving leachate from the columns.
The recirculation rate from the nitrification reactors to the column headspace was initially
set to 1 L/day. As in phase one, phase 2 was conducted under unsaturated conditions.
3.4 Physical Parameter Testing
3.4.1 Hydraulic Conductivity Testing
The hydraulic conductivity and the permeability of the porous media packed into the five
columns were measured before and after the experiment was run. To carry out these tests,
water was used at the beginning and leachate at the end. Leachate was used at the end
because of the possibility, if water was used, of perturbing or killing the bacterial
community already established in the columns. The flow discharge used to calculate the
initial and final hydraulic conductivity ranged from 130 to 320 mL/s. The differences in
energy (∆H) of every section (only represented by differences in pressure head) were
44
measured using the piezometers installed on the different ports of the columns. Tables 3.4
to 3.8 show the piezometric line of each column obtained during the tests carried out in
order to determine the original hydraulic conductivity.
Table 3.4 Composition of synthetic leachate (After Rowe et al, 2002)
Component Per 1L
Acetic (ethanoic) acid 7 mL
Propionic (propanoic) acid 5 mL
Butyric(butanoic) acid 1 mL
K2HPO4 30 mg
KHCO3 312mg
K2CO3 324 mg
NaCl 1440 mg
NaNO3 50 mg
NaHCO3 3012 mg
CaCl2 2882 mg
MgCl2.6H2O 3114 mg
NH4HCO3 2439 mg
CO(NH2)2 695 mg
Trace metal solutions (TMS)(se below) 1 mL
Na2S.9H2O Titrate to an Eh -120-180 mV
NaOH Titrate to a pH 5.8-6.0
Distilled water To make 1L
Component Per 1L
Composition of trace metal solutions (TMS)
FeSO4 2000 mg
H3BO3 50 mg
ZnSO4.7H2O 50 mg
CuSO4.5H2O 40 mg
MnSO4.H2O 500 mg
(NH4)6Mo7O24.4H2O 50 mg
Al2(SO4)3.16H2O 30 mg
CoSO4.7H2O 150 mg
NiSO4.6H2O 500 mg
96 % concentration H2SO4 (Anal R) 1 mL
Distilled water To make 1L
45
Tables 3.5-3.9 list the hydraulic conductivity values obtained for every section of the
column. (See Figure 3.1 Column schematic) The initial and final hydraulic conductivity
were measured for each section of the column under constant flow rate conditions, except
for the final hydraulic conductivity of the uppermost geotextile in each column. For the
tests, the columns were saturated and subjected to an upward flow of water (for the initial
hydraulic conductivity) or leachate (for the final hydraulic conductivity) and the
piezometers were read. With all this information, the hydraulic conductivity of the
different sections of the columns were calculated. For the uppermost geotextiles at the
end of the tests, the hydraulic conductivity was too low to enable this method to be used;
accordingly the hydraulic conductivity was measured using falling head tests carried out
on geotextile samples removed from the columns.
3.4.2 Porosity Testing
The porosity of all the porous media packed into the five columns was measured before
and after the laboratory program. The volume of each PVC column section was
determined and then packed with the porous media and filled with water. The liquid was
drained and its volume determined. The differences between the volume of the PVC
section and the drained liquid plus the additional water volume determined by drying the
porous media, is the volume of solids of the sample. With this information, the porosity
can be calculated by dividing the volume of voids by the total volume. (See From Figures
3.7 to 3.11 and Tables 3.10-3.14).
46
0
20
40
60
80
100
120
140
160
180
120 140 160 180
Total Head (cm)
Ele
vation (
cm
) Z
=0 a
t lo
west pie
zom
ete
r port
Figure 3.4 Column C1 initial piezometric line
Table 3.5 Hydraulic conductivity of different materials in column C1
Material Hydraulic Conductivity (cm/sec)
Geotextile 0.076
Stone 37.5-50mm 179
Geotextile 0.056
Stone 16-20mm 54
Stone 9-12.5mm + Stone 16-20 mm 21
Stone 9-12.5 mm 45
Geotextile 0.075
Stone 37.5-50mm 269
Q = 312 mL/sec
Area = 182.4 cm2
47
0
20
40
60
80
100
120
140
160
180
120 140 160 180
Total Head (cm)
Ele
vation(c
m)
Z=0 a
t th
e low
est pie
zom
ete
r port
Figure 3.5 Column CC1 initial piezometric line
Table 3.6 Hydraulic conductivity of different materials in column CC1
Material Hydraulic Conductivity (cm/sec)
Geotextile 0.04
Concrete 20-25mm 65
Geotextile 0.06
Stone 16-20mm 64
Stone 9-12.5mm + Stone 16-20 mm 22
Stone 9-12.5 mm 45
Geotextile 0.04
Concrete 20-25mm 161
Q = 186 mL/sec
Area = 182.4 cm2
48
0
20
40
60
80
100
120
140
160
180
120 140 160 180
Total Head (cm)
Ele
vation
(cm
) Z
=0
at th
e low
est p
iezom
ete
r po
rt
Figure 3.6 Column C2 initial piezometric line
Table 3.7 Hydraulic conductivity of different materials in column C2
Material Hydraulic Conductivity (cm/sec)
Geotextile 0.12
Stone 37.5-50mm 123
Geotextile 0.14
Stone 9-12.5mm 38
Stone 5-9mm + Stone 9-12.5 mm 17
Stone 5-9 mm 33
Geotextile 0.08
Stone 37.5-50mm 242
Q = 280 mL/sec
Area = 182.4 cm2
49
0
20
40
60
80
100
120
140
160
180
120 140 160 180
Total Head (cm)
Ele
vatio
n(c
m)
Z=
0 a
t th
e lo
we
st
pie
zom
ete
r po
rt
Figure 3.7 Column CC2 initial piezometric line
Table 3.8 Hydraulic conductivity of different materials in column CC2
Material Hydraulic Conductivity (cm/sec)
Geotextile 0.056
Concrete 20-25mm 112
Geotextile 0.14
Stone 9-12.5mm 75
Stone 5- 9mm + Stone 9-12.5 mm 32
Stone 5-9 mm 28
Geotextile 0.23
Concrete 20-25mm 112
Q =130 mL/sec
Area = 182.4 cm2
50
0
20
40
60
80
100
120
140
160
180
120 140 160 180
Total Head (cm)
Ele
vation (cm
) Z=0 a
t th
e lowest pie
zom
ete
r port
Figure 3.8 Column C3 initial piezometric line
Table 3.9 Hydraulic conductivity of different materials in column C3
Material Hydraulic Conductivity (cm/sec)
Geotextile 0.08
Stone 37.5-50mm 241
Geotextile 0.15
Stone 9-12.5mm 20
Stone 5-9mm + Stone 9-12.5 mm 16
Stone 5-9 mm 21
Geotextile 0.085
Stone 37.5-50mm 242
Q = 280 mL/sec
Area = 182.4 cm2
51
0
20
40
60
80
100
120
140
160
180
0 0.2 0.4 0.6 0.8 1
Porosity
Ele
va
tio
n (
cm
)
Figure3.9 Initial porosity profile of column C1
Table 3.10 Column C1 initial porosity
Material Porosity
Geotextile 0.90
Stone 37.5-50mm 0.42
Geotextile 0.90
Stone 16-20mm 0.41
Stone 9-12.5mm + Stone 16-20 mm 0.39
Stone 9-12.5 mm 0.38
Geotextile 0.90
Stone 37.5-50mm 0.42
52
0
20
40
60
80
100
120
140
160
180
0 0.2 0.4 0.6 0.8 1
Porosity
Ele
va
tio
n (
cm
)
Figure 3.10 Initial porosity profile of column CC1
Table 3.11 Column CC1 initial porosity
Material Porosity
Geotextile 0.90
Concrete 20-25mm 0.44
Geotextile 0.90
Stone 16-20mm 0.41
Stone 9-12.5mm + Stone 16-20 mm 0.39
Stone 9-12.5 mm 0.38
Geotextile 0.90
Concrete 20-25mm 0.44
53
0
20
40
60
80
100
120
140
160
180
0 0.2 0.4 0.6 0.8 1
Porosity
Ele
va
tio
n (
cm
)
Figure 3.11 Initial porosity profile of column C2
Table 3.12 Column C2 initial porosity
Material Porosity
Geotextile 0.90
Stone. 37.5-50mm 0.42
Geotextile 0.90
Stone. 9-12.5mm 0.38
Stone. 5-9mm + Stone. 9-12.5 mm 0.37
Stone. 5-9 mm 0.36
Geotextile 0.90
Stone.37.5-50mm 0.42
54
0
20
40
60
80
100
120
140
160
180
0 0.2 0.4 0.6 0.8 1
Porosity
Ele
va
tio
n (
cm
)
Figure 3.12 Initial porosity profile of column CC2
Table 3.13 Column CC2 initial porosity
Material Porosity
Geotextile 0.90
Concrete 20-25mm 0.44
Geotextile 0.90
Stone 9-12.5mm 0.38
Stone 5- 9mm + Stone.9-12.5 mm 0.37
Stone 5-9 mm 0.36
Geotextile 0.90
Concrete 20-25mm 0.44
55
0
20
40
60
80
100
120
140
160
180
0 0.2 0.4 0.6 0.8 1
Porosity
Ele
vation (
cm
)
Figure 3.13 Initial porosity profile of column C3
Table 3.14 Column C3 initial porosity
Material Porosity
Geotextile 0.90
Stone 37.5-50mm 0.42
Geotextile 0.90
Stone 9-12.5mm 0.38
Stone 5-9mm + Stone 9-12.5 mm 0.37
Stone 5-9 mm 0.36
Geotextile 0.90
Stone 37.5-50mm 0.42
56
3.5 Water Quality Testing Program
Water quality testing was performed on samples of influent and effluent leachate at a
different frequency according to the experiment phase (see Table 3.15). Tests were
conducted to obtain: pH, Electrical Conductivity (Ec), Oxidation Reduction Potential
(Eh), Ammonia, Nitrite, Nitrate, Total Nitrogen, Alkalinity, Total Organic Carbon
(TOC), Volatile Fatty Acids (VFA), Calcium Hardness, Chemical Oxygen Demand
(COD), Biological Activity Reaction Test (BARTTM), Sodium, Magnesium, Sulphate,
Iron, Chloride. The frequency of sampling conducted in each phase of the experiment is
listed in Table 3.15.
Table 3.15 Frequency of the water quality testing program
Parameter Frequency 1st phase Frequency 2
nd phase
pH Three times/week Three times/week
Ec Three times/week Three times/week
Eh Twice/week Twice/week
Ammonia Twice/week Twice/week
Nitrite Not performed Twice/week
Nitrate Not performed Twice/week
Total Nitrogen Infrequently Once/week
Alkalinity Infrequently Twice/week
Calcium Hardness Once/week Once/week
Chemical Oxygen Demand Once/week Once/week
Biological Activity Reaction
Tests (BARTTM),
Once/week Infrequently
Total organic Carbon (TOC) Infrequently Infrequently
Total Inorganic Carbon
(TIC)
Infrequently Infrequently
Volatile Fatty Acids (VFA), Infrequently Infrequently
Sodium Infrequently Infrequently
Magnesium Infrequently Infrequently
Sulphate Infrequently Infrequently
Iron Infrequently Infrequently
Chloride Infrequently Infrequently
57
pH, Electrical conductivity (Ec) and Oxido Reduction Potential (Eh) were measured with
a HACH 44600TM probe. The calibration of the probe was checked from time to time
using standard solutions.
Ammonia was measured with TECATOR TM distillation apparatus . Nitrite and Nitrate
with an auto-analyser by TechniconTM, Total nitrogen and COD were monitored using the
HACH TM Reactor Digestion Method. This method uses vials that are analysed with the
HACH TM DR 4000U Spectrophotometer.
Alkalinity was measured using an autotitrator and bringing the pH down to 8.3 for
carbonate alkalinity and down to 4.5 for bicarbonate alkalinity. The machine used was
pH titrator E512 Metrohm Herisau. Sodium and Potassium were measured with a
Corning flamephotometer 430. Iron with a Peekinelmer 5000 atomic absorption
spectrophotometer. Sulphates were measured using the gravimetric method with ignition
of residue.
TOC (Total Organic Carbon) and TIC (Total Inorganic Carbon) analyses were performed
with a Telemar DorhmanTM , Phoenix 800t uv-persulfate analyser. Volatile Fatty Acids
(VFA) was carried out by distillation under standard methods.
Biological Activity Reaction Tests (BARTTM) were used to monitor the presence and
changes to the microbial community within the columns and it is performed with
prepared vials designed to measured bacterial activity.
58
3.6 Gas Production
The volume of gas generation during phase one was measured occasionally using a GEM
2000 device which indicates the percent of volume CO2, CH4, O2 and Balance (which
represent any element different to the three previously mentioned). For phase 2, gas
production was collected in 1L Tedlar bags connected to the gas headspace at the top of
each column. The Tedlar bags were almost emptied (some gas was left inside the bag in
order to avoid any vacuum being created by sampling. This ensured that the gas pressure
inside the columns would always be known, i.e. atmospheric. Similarly, the bags were
emptied before they could fill completely and cause a build-up of pressure inside the
columns. Cumulative gas production was measured once the experiment stabilized.
Results are presented in Chapter Four. Analysis of the gas concentration was performed
at the Saskatchewan Research Center Analytical Laboratory.
3.7 Simulation of Leachate Generation and Recirculation
One of the objectives of this thesis was to create operating conditions similar to those that
occur in real landfills in order to evaluate the efficiency in leachate collection.
Infiltration into North American landfills may range from 0.05 m3/m
2/yr in dry regions to
0.2 m3/m
2/yr in humid regions (Quian et al, 2002). Assuming an average rate of
infiltration into eastern Canadian landfills of 0.175 m3/m
2/yr, a medium size landfill of 25
ha would produce approximately 120 m3/day of leachate.
59
Furthermore, assuming a typical recirculation rate of 5 times the leachate production rate
(in this case 600 m3/day) and a typical hydraulic retention time (HRT) of 2.7 days for the
nitrification reactor; then the volume of the nitrification reactor needed would be
approximately 1,600 m3. This detention requirement could easily be met using concrete
tanks or a small lined lagoon at the landfill site. These field estimates were taken into
account in scaling the laboratory tests; however if those field parameters were used in the
experiment, the leachate production rate would be rather low for practical parameter
testing (approximately 9 mL/day) due to the small cross sectional area of the columns
(182.4 cm2).
As a result of the previous considerations, the synthetic leachate input or infiltration rate
was set at 100 mL/day (roughly 10 times larger than the midsize Canadian landfill
previously discussed). The recirculation rate was also initially set 10 times larger that the
infiltration rate. The nitrification reactor HRT was maintained at 2.7 days. These
operating conditions resulted in a nitrification reactor volume of 3 L and a recirculation
rate of 1 L/day for the experiment.
All the experiment operating parameters were subject to adjustments depending on the
results obtained upon implementation of the feed of synthetic leachate into the columns.
However, the experiment attempted as much as possible to maintain operating conditions
representative of typical landfill conditions.
60
Chapter 4 Evaluation of Leachate Quality Using a Modified Drainage LCS
4.1 Introduction
The concept of a bioreactor landfill, with leachate recirculation as its main feature, has
the advantage of being effective and economically viable mainly because it relies entirely
upon in-situ biological treatment. However, the utilization of landfills for treatment in
addition to accumulation of refuse has the potential to create additional geotechnical and
environmental challenges. These include slope failure due to decreases in internal and/or
interface shear strength (Quian et al., 2002), clogging of the leachate collection
systems(LCS) (Brune et al., 1991; Fleming et al., 1999, 2004 and VanGulck et al. 2003)
and increased levels of ammonia (Pohland, 1995).
This chapter describes the results obtained in synthetic leachate COD, nitrogen and
calcium concentration as well as the composition of the biogas generated, after synthetic
leachate underwent anaerobic and aerobic biodegradation through the use of columns
simulating leachate collection drains and external aerated reactors, respectively. A mass
balance for carbonaceous and nitrogenous compounds is also performed in order to
confirm the leachate treatment effectiveness.
61
4.2 Laboratory Testing Results
4.2.1 pH and Alkalinity.
Figure 4.1 and 4.2 show the pH of the anaerobic column discharge and the aerated
nitrification reactor effluent (termed the aerated effluents). Considering that the synthetic
leachate has an initial pH of 6 (see Table 4.1), the much higher values of 8.3-8.6 in the
anaerobic column discharge, suggest a high bacterial population that rapidly decreases
the volatile organic acid content (mainly represented by acetic acid) of the leachate.
Figure 4.1 also shows that the pH in the columns packed with crushed concrete is slightly
higher than for those columns packed with natural gravel (this subject is deepened in
section 5.3). This effect is not apparent in the aerated effluent as shown in Figure 4.2.
Figures 4.1 and 4.2 also show that the pH of the anaerobic column discharge is about
one-half of a pH unit lower than the aerated effluent, reflecting the evolution of dissolved
carbonates from the open vessel of the nitrification reactor. Thus the anaerobic
degradation of the leachate in the column replaces the weak carboxylic acids with even
weaker carbonic acid (Fleming et al., 1999). When the leachate is discharged to the
aerated nitrification reactor, off gassing of dissolved CO2 occurs readily.
This CO2 off gassing also explains the difference between alkalinities as seen in Figures
4.3 and 4.4. The biogas produced anaerobically in the columns was measured to be 1.6%
62
8
8.5
9
9.5
0 10 20 30 40 50 60 70 80 90 100
Elapsed time (days)
pH
pH C1 pH CC1
pH C2 pH CC2
pH C3
Figure 4.1 pH in anaerobic column discharge
63
8
8.5
9
9.5
0 10 20 30 40 50 60 70 80 90 100
Elapsed time (days)
pH
pH C1 pH CC1
pH C2 pH CC2
pH C3
Figure 4.2 pH in aerated effluent
64
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
0 10 20 30 40 50 60 70 80 90 100
Elapsed time (days)
Alkalinity mg/L as CaCO3
Alkalinity C1 Alkalinity CC1
Alkalinity C2 Alkalinity CC2
Alkalinity C3
Figure 4.3 Alkalinity in anaerobic column discharge
65
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
0 10 20 30 40 50 60 70 80 90 100
Elapsed time (days)
Alkalinity mg/L as CaCO3
Alkalinity C1 Alkalinity CC1
Alkalinity C2 Alkalinity CC2
Alkalinity C3
Figure 4.4 Alkalinity in aerated effluent
66
CO2, thus within the columns reactors, leachate is assured to be at equilibrium with a CO2
partial pressure (0.016 atm) higher than in the aerated nitrification reactor (0.0003 atm).
This condition makes leachate in the column reactor more acidic (by carbon dioxide
hydrolysis) and at the same time increases the alkalinity due to carbonates augmentation.
4.2.3 Ammonia Conversion to Nitrate
Figure 4.5 presents test results for ammonia concentration in the outflow of the aerated
nitrification reactors. The ammonia removal efficiency is referenced to the same source
concentration, that of the synthetic leachate. During the first 35 days some operational
difficulties resulted in high levels of ammonia. This was solved with minor changes to the
system that enabled a more continuous flow. Figures 4.5 and 4.6 show that after 50 days a
stable nitrification process was established and all ammonia was converted to NO3-.
4.2.4 Nitrate Conversion to Nitrogen Gas
Data from only a single column (C2) will be presented in this section given that all the
columns have followed very similar trends for each of the parameters tested. The other
columns show essentially the same behaviour.
Figure 4.7 shows that after day 50, stable conditions are evident, ammonia is completely
oxidized and only NO3- dominates the nitrogenous compounds. Further, Figure 4.8 does
67
0
100
200
300
400
500
600
700
800
900
0 10 20 30 40 50 60 70 80 90 100
Elapsed time(days)
Ceff/Co Ammonia C1
Ammonia CC1Ammonia C2Ammonia CC2ammonia C3Co
Co = 766 mg/L as N
Figure 4.5 Ammonia concentration in aerated effluent
68
0
50
100
150
200
250
0 10 20 30 40 50 60 70 80 90 100
Elapsed time(days)
NO3 mg/L as N
C1 CC1
C2 CC2
C3
Figure 4.6 NO3- concentration in the aerated effluent
69
not show any sizable amount of either nitrite or nitrate in the anaerobic column discharge.
This suggests nitrate conversion to N2(gas). Nitrogen gas production was confirmed by
collection analysis of gas produced by the column. Approximately 37% of the daily gas
production was N2(gas).
It is also interesting to see that between days 37 and 50, NO2- was the main nitrogenous
compound produced in the aerated nitrification reactor (See Figure 4.7). However figure
4.8 shows almost no presence of it in the column reactor. This could be explained by
anaerobic ammonium oxidation (Anammox) where ammonium is oxidized under
anaerobic conditions, NO2- acts as a final electron acceptor and CO2 is used as the main
carbon source for growth (Jetten et al ,2001). The stoichiometry of this reaction is:
{1} NH4+ + NO2
- → N2 + 2H2O
The loss of ammonia by this process requires approximately equimolar concentrations of
ammonia and nitrite (Price et al., 2004) and normally occurs at the interface of an
anaerobic/aerobic system (Schmidt et al., 2002). Both conditions are met in this
experiment. Nonetheless, further investigation, beyond the scope of this study, is
necessary to determine the validity of this hypothesis.
70
0
100
200
300
400
500
600
700
800
900
0 10 20 30 40 50 60 70 80 90 100
Elapsed time(days)
Nitrogen (mg/L)
NH3
NO2
NO3
Total N
Co
Co = 766 mg/L as N
Figure 4.7 Nitrogen species in aerated effluent (column C2)
71
0
100
200
300
400
500
600
700
800
900
0 10 20 30 40 50 60 70 80 90 100
Elapsed time (days)
Nitrogen(mg/L )
NH3
NO2
NO3
Total N
Co
Co = 766 mg/L as N
Figure 4.8 Nitrogen species in anaerobic column discharge (column C2)
72
4.2.5 Total Nitrogen and Total Nitrogen Removal
Figure 4.9 and 4.10 clearly show that nitrogen is removed from the system. The total
nitrogen present in both the aerated effluent and the anaerobic column discharge are
significantly reduced compare to the leachate input. Total nitrogen decreased from 766
mg/l to 130 mg/l as N (as an average), resulting in approximately 84% removal.
During the first 40 days of the experiment, there is some apparent loss of nitrogen
between the column discharge and the nitrified effluent likely reflecting volatilization of
ammonia from the aerated nitrification tank. This effect seems to have diminished after
about 45 days as the main consequence of the improving conversion of ammonia in the
reactor.
4.2.6 COD Removal
Chemical Oxygen Demand (COD) in this laboratory test is represented by the carboxylic
acids present in the synthetic leachate (acetic, propionic and butyric). As shown in
Figures 4.11 and 4.12, COD removal was stable and achieved approximately 97%
removal with the exception of a brief upset during week 3. This clearly demonstrates and
confirms that leachate collection drains can be utilized as an attached growth bioreactor
for degradation of organic compounds in MSW leachate
73
0
100
200
300
400
500
600
700
800
0 10 20 30 40 50 60 70 80 90 100
Elapsed time(days)
Tota
l N
mg/L
C1
CC1
C2
CC2
C3
Total Nitrogen in SyntheticLeachate
Figure 4.9 Total nitrogen in anaerobic column discharge
74
0
100
200
300
400
500
600
700
800
0 10 20 30 40 50 60 70 80 90 100
Elapsed time(days)
Tota
l N
mg/L
C1
CC1
C2
CC2
C3
Total Nitrogen in Synthetic Leachate
Figure 4.10 Total nitrogen in aerated effluent
75
Previous studies conducted with attached growth reactors simulating LCS showed a COD
removal of approximately 14,000 mg/L as O2 (from 16,000 mg/L to 2,000 mg/L as O2)
(Fleming et al., 1999; Fleming and Rowe 2004). In those studies the remaining COD was
considered to be refractory materials that were difficult to biodegrade. Armstrong, (1998)
and Rowe et al., (2002) also showed that one important factor in COD depletion is
particle size of the porous medium. This finding is supported by the experiment. The
geotextiles in the columns have a large a surface area and porosity and held a large
bacterial population. Hence, the geotextile functions as an excellent fixed biofilm reactor
that is largely responsible for the leachate treatment (see chapter 5.1).
The Hydraulic Retention Time of the aerated nitrification tank (HRT) also plays an
important role in the removal of contaminants. The experiment started at a HRT of 2.7
days which produced a removal efficiency of approximately 90%. From day 14 on, the
HRT was increased to 5 days. Figure 13 shows that the removal efficiency increased by
7% with a HRT of 5 days. Later on, around day 80, the HRT was cut back to 3.5 days for
a week and an increase in COD and disruption of the aerated nitrification reactors were
observed. Because of these results, the HRT was moved back to 5 days and the
experiment returned to previous satisfactory performance.
76
0.00
0.20
0.40
0.60
0.80
1.00
0 10 20 30 40 50 60 70 80 90 100
Elapsed time(days)
Ceff/Co
C1 CC1
C2 CC2
C3 Co
Co = 17,345 mg/l as O2
Figure 4.11
4.11 COD removal in anaerobic column discharge
77
0.00
0.20
0.40
0.60
0.80
1.00
0 10 20 30 40 50 60 70 80 90 100 110 120
Elapsed time(days)
Ceff/Co
C1 CC1
C2 CC2
C3 Co
Co = 17,345 mg/l as O2
HRT
2.7
days
HRT
5
days
HRT
3.7
days
HRT
2.7
days
Figure 4.12 COD removal in aerated effluent (the HRT values refer to the aerated nitrification tank only)
78
The most likely reason for this behaviour lies in the nature of the experiment. Higher
recirculation flow rates will transport more oxygen from the aerated reactor into the
anaerobic column reactor, thus inhibiting the mainly methanogenic bacteria that are
strictly anaerobes.
Overall, the synthetic leachate COD was reduced from approximately 17,000 mg/L to
600 mg/L. The following may have contributed to increased COD removal in this system
compared to Fleming et al.(1999) and Fleming and Rowe (2004):
1. The presence of nitrifying and de-nitrifying bacteria in addition to the heterotrophic
population of bacteria that utilize organic matter as substrate, and
2. The coupled nitrification reactor may result in additional degradation of organic
compounds. Close examination of the data presented in Figure 4.11 and 4.12 suggests
that there is not significant COD removal in the nitrification reactor itself, but that the
nitrified effluent may undergo additional conversion when recycled through the anaerobic
column. It is not known whether this effect reflects lessening the degree of inhibition by
ammonia as the ammonia concentration is decreased within the column or some other
synergistic effect.
79
4.2.7 Gas Production
Gas was collected in Tedlar bags at the top of the columns. Figure 4.13 shows a steady
gas production at 180 mL/day for three of the column reactors
The biogas produced in the columns contains a mixture of methane, nitrogen and CO2.
Samples taken for analysis have confirmed approximately 60% methane, 1.6% CO2 and
37% nitrogen. This gas composition is substantially different than that typically found in
landfill gas. If it were possible to sustain this gas composition in the field, there could be
beneficial consequences such as less acidic condensate.
Table 4.1 Gas composition
Gas CH4 N2 CO2 O2
% by volume 59-61 36-39 1.0-1.6 < 1.0
Although several studies (Price et al. 2003, Onay and Pohland, 1998 and Lin and Chen,
1995) found that methane yields decrease when denitrification takes place. It can be
argued that those findings do not necessarily fully apply for this experiment and even for
a real landfill. The reason for decreased methane production lies in the fact that bacteria
obtain much more energy converting nitrite to nitrogen gas than converting carboxylic
acids to methane ( ∆G˚ = -1120 KJ and ∆G˚ = -31 KJ respectively). Price et al (2003)
used up to 5 moles of nitrate (310,000 mg/L as NO3-) and a BOD of 10,000 mg/L as O2.
80
0
2000
4000
6000
8000
10000
12000
14000
0 20 40 60 80 100
Elapsed Time (days)
Cummulative Gas Production (ml)
Cum.gas CC1
Cum.Gas C2
Cum.Gas CC2
Figure 4.13 Anaerobic column reactors gas production
81
At such a large nitrate/BOD ratio, denitrifying bacteria will out compete
methanogenic bacteria. However, in a relatively new landfill (as this experiment
simulates) that is not the case.
The supply of organic carbon represented by a COD of 17,000 mg/L as O2 proves to
be enough to treat nitrogen values around 760 mg/L as N with out either affecting
methane production or creating dissimilatory nitrate reduction to ammonia (DNRA)
which may occur at low N/COD ratio and reduces nitrate back to ammonia (Tiedje,
1988).
4.3 Mass Balance Analysis
Mass balance calculations were conducted in an attempt to account for all nitrogen
and carbon in the treatment system. The mass balance calculations provide
quantification of the conversion processes involving nitrogen and carbon compounds
and confirm the reported treatment efficiencies.
4.3.1 Nitrogen Mass Balance
The nitrogen mass balance calculations are summarized in Table 2. Nitrogen enters
the system in the synthetic leachate (source) mainly as ammonium bicarbonate and
urea. The nitrogen mass input is 77 mg/day based upon an average measured
leachate concentration of 766 mg/L as N and an input flow of 0.1 L/day.
82
Table 4.2 Nitrogen mass balance
Component Phase Nitrogen Form Mass/day
(mg/day)
Source
Synthetic leachate inflow aqueous ammonia and urea 77
Σ Source = 77
Sinks
Reactor effluent outflow aqueous nitrate 14
Column off gassing gaseous nitrogen gas 74
Σ Sinks = 88
(Σ Sinks - Σ Source)/ Σ Source = 14%
Ammonia nitrogen is converted to nitrate in the nitrification reactor, and recycled
nitrate is converted to nitrogen gas by denitrification in the column. Therefore,
nitrogen leaves the system (sinks) as nitrogen gas from the column, and as nitrate in
the effluent flow from the nitrification reactor. The measured nitrogen content of the
off-gas collected from the column was 37% by volume. The nitrogen gas nitrogen
mass outflow (74 mg/day) was estimated using nitrogen mass density at 35°C, an
off-gas pressure of one atmosphere, and the measured total off-gas flow (180
mL/day). The nitrate nitrogen mass outflow (14 mg/day) was estimated using the
measured nitrate concentration in the nitrification reactor (140 mg/L) and the
effluent flow (0.1 L/day). Therefore, the total estimated nitrogen outflow (sink) is
88 mg/day.
83
The nitrogen mass balance source and sinks estimates agree within approximately
14%. This small discrepancy is likely due to sampling and analysis errors. Despite
this small error, the calculations clearly illustrate the efficient conversion and
removal of ammonia nitrogen from the leachate entering the treatment system.
4.3.2 Carbon Mass Balance
Carbon enters the system mainly as volatile fatty acids (VFA) in the synthetic
leachate (source). The VFA undergo anaerobic biodegradation in the column and
are converted to methane (CH4), carbon dioxide (CO2) and other minor by-products.
The CH4 and CO2 are both off-gassed from the column. The conversion of VFA
results in the removal of aqueous phase COD and a significant increase in pH due to
a dramatic increase in carbonate alkalinity resulting from CO2 production and
dissolution. The increased carbonate alkalinity and pH cause precipitation of
carbonate solids (primarily CaCO3) within in the column. The leachate leaving the
column and entering nitrification reactor is supersaturated with CO2 for open to
atmosphere conditions. Therefore, off gassing of CO2 from the nitrification reactor
is an additional loss of carbon from the system. Each of the carbon sinks is
estimated in the carbon mass calculations.
The carbon mass balance calculations are summarized in Table 3. The carbon input
(source) is 600 mg/day based upon an average measured leachate concentration of
84
6000 mg/L as C and an input flow of 0.1 L/day. The input carbon concentration was
measured using a total carbon analyser.
Table 4.3 Carbon mass balance
Component Phase Carbon Form Mass/day
(mg/day)
Source
Synthetic leachate inflow Aqueous Carboxilic acids,
Carbonates
600
Σ Source = 600
Sinks
Reactor effluent outflow Aqueous Organic carbon 6
Aqueous Inorganic carbon 116
Deposition of Carbonate
precipitates
Solid Calcite precipitate 33
Column off gassing
Gaseous Carbon dioxide 1
Gaseous Methane 51
Nitrification Reactor off
gassing
Gaseous Carbon dioxide 102
Σ Sinks = 309
(Σ Sinks - Σ Source)/ Σ Source = 49%
Carbon mass flow to each of the carbon sinks was estimated as follows:
• The carbon outflow in the reactor effluent (122 mg/day) was estimated using
the measured inorganic and organic carbon content of the effluent (1,160 and
60 mg/L respectively) and the effluent flow (0.1 L/day).
85
• The measured CH4 and CO2 content of the off-gas collected from the column
was 60% and 1.6% respectively by volume. The carbon mass outflow rates
for CH4 and CO2 (51 and 1 mg/day respectively) were estimated using CH4
and CO2 mass density at 35°C, an off-gas pressure of one atmosphere, and
the measured total off-gas flow (180 mL/day).
• The loss of carbon to deposition of carbonate precipitates in the column was
estimated using the drop in calcium concentration between the leachate and
effluent outflow assuming the vast majority of the carbonate solids are
CaCO3, and that the dominate carbonate species in the effluent is
bicarbonate. Using these simplifying assumptions, the measured drop in
calcium concentration from 2900 to 150 mg/L through the system, and an
effluent flow rate of 0.1 L/day produced an estimate of 33 mg/day carbon
loss to precipitates in the column.
• The loss of carbon due to off gassing of CO2 from the nitrification reactor
can be estimated using the change in total alkalinity between the inlet and
outlet of the reactor (a drop from 5,700 mg/L to 5,000 mg/L as CaCO3).
Assuming the dominate alkalinity species is bicarbonate (based upon pH)
and the alkalinity lost is converted to CO2, then the carbon loss from the
system can be calculated using the reactor flow of 0.6 L/day (sum of the
leachate and re-circulation flow). The estimated carbon loss using this
approach is 102 mg/day.
86
The total carbon loss and deposition represented by the sinks described above is 309
mg/day.
The carbon mass balance calculations indicate a large proportion of the carbon
entering the system (~ 50%) is unaccounted for. However, at least two additional
carbon sinks exist that have not been quantified. First, a significant amount of CO2
may be off gassed from the column discharge sample collection apparatus. This
CO2 loss is not accounted for in the calculation approach described above. Second,
there is likely a build up of biomass within the column as the experiment progresses.
Whereas the active biomass may approach an equilibrium condition, inactive
biomass may continue to accumulate as biofilm thickens and cells die-off due to
restricted access to substrate.
Despite the lack of closure of the carbon mass balance the calculations provide very
encouraging results with respect to organic carbon discharged from the system in the
aqueous phase. Of the 600 mg/day organic carbon entering the system in the
leachate, only 60 mg/day organic carbon leaves the system in the nitrification reactor
effluent. These calculations confirm the high level of treatment efficiency provided
by the system that was indicated by the COD measurements presented earlier.
87
4.4 Conclusions
The experimental set-up simulated a leachate collection system (LCS) with
recirculation coupled with a nitrification reactor. The results of the experiment
showed that in-situ passive treatment of the leachate COD and ammonia is feasible.
Furthermore, the tests shown that the LCS can be used as an efficient fixed biofilm
reactor. The large surface area of the geotextile allows microbes to attach and grow.
The biofilm that develops under anaerobic conditions, biodegrades carboxylic acids,
producing mainly carbonates and methane gas as by products.
The generation of carbonates precipitate can be a source of incrustation and
obstruction of the drain and/or leachate collection pipes. Saturation of the leachate
with CO32- causes increased pH, and precipitates Ca
2+ as calcium carbonate that has
the potential to clog the system. Clogging then, may result due to CO2 released as a
product of organic matter degradation and carbonate system equilibrium with the
surrounding environment.
Nitrification and denitrification are processes that can be achieved reliably in the
coupled LCS and aerated reactor system. The treatment of 776 mg/L of nitrogen as
N (in the form of urea and ammonium bicarbonate) was carried out with an 84%
average efficiency. Hydraulic retention time (HRT) plays a fundamental role in the
system. The optimum HRT for the experiment was 5 days. Changes in this
parameter led to inefficiencies in both ammonia and COD depletion. The pH values
88
measured in the experiment were higher than values for optimal microbial growth (7
to 8 for denitrification and 7.5 to 8.6 for nitrification (Shiskowsky and Mavinik,
1995)). This shows the denitrifying and nitrifying bacteria are resilient.
COD removal (compared to the leachate input concentration) reached levels of up to
95% in the column reactor effluent and 97% in the nitrification reactor. Previous
studies by Fleming and Rowe (2004) showed the difficulty of treating organic matter
using suspended and porous medium beyond the 2000 mg/L concentration. This
was thought to be due to the presence of recalcitrant organic compounds that are
hard for bacteria to utilize. In this study COD levels were reduced to below 600
mg/L. Synergistic effects within the column may explain the lower effluent
concentrations observed. The most likely effect is denitrifying bacteria and
methanogenic bacteria competing for organic carbon as an energy source. As a
result, a higher proportion of the available organic carbon is utilized, even the
recalcitrant compounds.
The biogas produced in the columns was comprised of methane (60%), nitrogen
(37%) and CO2 (1.6%). The nitrogen and carbon dioxide values are substantially
different than those typically produced in a landfill. The CO2 content of the biogas in
particular, is much lower than typically produced in a landfill. The consequences of
a reduced CO2 level are beneficial. A low carbon dioxide value decreases the
amount of carbonates in the leachate, reducing the potential risk of clogging. It also
89
means a less acidic (corrosive) environment for a landfill gas collection network,
thus extending its serviceable life.
The overall results obtained from the experiment show that in-situ treatment at
landfills site is viable. The required HRT for the aerated nitrification reactor is small
enough to make the system physically and economically feasible. Furthermore,
despite the fact that methanogenic and denitrifying bacteria compete for carbon
availability, it is reasonable to think that the amount of refuse disposed of in a
landfill will be able to provide enough organic carbon to feed both microbial
populations allowing COD to be depleted and nitrate to be off-gassed. As a result,
methane will be produced in amounts that would be attractive for use as a source of
energy.
90
Chapter 5 Hydraulic Performance of the Modified Drainage LCS
5.1 Introduction
In order for a landfill LCS to work as a fixed biofilm reactor, it has to develop a microbial
community within its porous media. Since geotextiles and drains are integral parts of a
landfill LCS, their appropriateness for biofilm growth has to be assessed. Rollin and
Lombard (1988), Koerner and Koerner (1992), and Mlynarek and Vermeersch (1999)
have demonstrated the suitability of geotextiles for supporting media growth of bacterial
biofilm. This is mainly due to their large surface area and porosity, which facilitates
microbial attachment and growth. Correspondingly, Armstrong (1998) permeated
leachate through a porous media packed in several columns resembling landfill LCS
drains, and found a direct relationship between bacterial growth (leachate treatment) and
the surface area of the drain material.
This chapter analyzes the impact on the hydraulic properties of the laboratory columns
after being operated as fixed biofilm reactors with all the physical, chemical and
biological consequences that such a mode of operation entails. It also projects the
hydraulic performance findings of the laboratory experiment to a full scale landfill,
represented by a typical North American landfill.
91
Lastly, the performance of the two columns packed with recycled crushed concrete is
analyzed, and compare to the three columns packed with natural gravel. The purpose of
this analysis is to investigate the potential (and benefits) of using a recycled material, in
this case concrete, as a drainage material in a LCS.
5.2 Effects of leachate treatment in Geotextiles
Table 5.1 shows the coefficient on hydraulic conductivity of the different sections of the
columns at the beginning and at the end of the experiment. The hydraulic conductivity
results were obtained using a constant head method with flow rates ranging from 130 to
310 mL/day. For the uppermost geotextile a falling head method was used.
Except for the uppermost geotextiles (section F1), the sections showed no meaningful
change in hydraulic conductivity before and after the experiment. As an average, the
decrease in the hydraulic conductivity of the uppermost geotextiles was approximately
1.2 orders of magnitude. It went from an average of 0.05 cm/s to an average of 0.0028
cm/s
Figure 5.1 shows the piezometric profiles of the columns at the beginning and at the end
of the experiment. As can be seen, no significant difference was found before and after
the experiment, except for the uppermost geotextile, which showed a considerable
increase in pressure head across the uppermost geotextile filter.
92
Table 5.1 Hydraulic conductivity of the different sections of the columns C1, C2, C3, CC1 and CC2
Column C1 Column C2 Column C3 Column CC1 Column CC2
Section
K Ini.
(cm/sec)
K Fin.
(cm/sec)
K Ini.
(cm/sec)
K Fin.
(cm/sec)
K Ini.
(cm/sec)
K Fin.
(cm/sec)
K Fin.
(cm/sec)
K Fin.
(cm/sec)
K Ini.
(cm/sec)
K Fin.
(cm/sec)
F1 7.6E-2 2.5E-3 1.2E-1 2.1E-3 8E-2 5.7E-3 4E-2 2.6E-3 5.6E-2 1.7E-3
S2 179 180 123 123 241 193 65 62 112 112
F2 5.6E-2 5.3E-2 1.4E-1 1.3E-1 1.5E-1 1.4E-1 6E-2 6E-2 1.4E-1 1.5E-1
S3 54 56 38 38 20 20 64 54 75 75
S4 45 45 17 17 21 21 45 45 28 28
F3 7.5E-2 7E-2 0.8E-2 8.4E-2 8.5E-2 8.6E-2 4E-2 4E-2 2.3E-1 2.7E-1
S5 270 270 241 121 242 194 161 161 112 112
93
The decrease in hydraulic conductivity of the uppermost geotextiles was a result of
several factors including; particles washed out of the overlying wet refuse, bacterial slime
formation, and calcite precipitation. All the uppermost geotextiles presented a “muddy”
appearance like the one (column C2) shown in Figure 5.2.
0
20
40
60
80
100
120
140
160
180
140 150 160 170 180 190 200 210 220
Total Head (cm)
Elevation (cm) Z=0 at the lowest piezometer port
C1 ini.
C2ini
C3ini
CC1ini
CC2ini
C1fin
C2fin
C3fin
CC1fin
CC2fin
Figure 5.1 Initial and final piezometric line of the columns
94
Figure 5.2 Uppermost geotextile in column C2 after 100 days of operation
Figures 5.3 and 5.4 show vertical thin section photomicrographs of the uppermost needle
punched non-woven geotextile before and after the experiment. Mineral deposits,
partially responsible for the lower hydraulic conductivity, can be seen as white blotches
in Figure 5.4.
Geotextile
95
Figure 5.3 Geotextile before the experiment
Figure 5.4 Geotextile after the experiment
96
5.2.2 Clog Material Analysis of Geotextiles
A gravimetric analysis was carried out in order to determine the composition of the
material within the uppermost geotextiles (i.e. the relative proportion of volatile solids,
calcite, and other minerals or “refractory material”). Two samples with an area of 32 cm2
were cut from the uppermost geotextile of each column and were weighed in an oven-
dried condition; following ignition at 550 ˚C and finally after combustion at 900˚C. The
average mass of a clean geotextile was determined by weighting 30 pieces of clean
geotextile of equal area (32 cm2). The average mean mass and standard deviation were
found to be 1.19 g and 0.14 respectively. The geotextiles samples were subsequently
weighed after ignition in a laboratory furnace at 550˚C and it was found that the
polypropylene material was almost completely burned off; LOI (loss on ignition) was
99.6 % at 550˚C. Therefore, if the area of the pieces of the uppermost geotextiles is
known, its mass can be reasonably estimated within a range of (+-) 0.69 units. This
estimate of mass is required when determining the mass of organic and inorganic material
attached to or retained within the fabric. The results of this analysis are listed in Table
5.2.
Fleming et al. (1999) performed elemental analysis of slime and solid samples removed
from Canadian landfill drains. Calcite was found to comprise between 40 and 50 % of the
dry weight of the clog material that was deposited within the drains.
97
Table 5.2 Classification of clog material within the uppermost geotextile
Measured Calculated Dry Wt basis
Column M 105˚C M 550˚C M 900˚C Mvs(g) Mcalc(g) Mrefr(g) total mass(g) % VS % Calc % Refr
C1 1.7558 0.55 0.4209 0.01 0.29 0.26 0.56 2.10 52.23 45.67
C1 1.8674 0.565 0.3772 0.11 0.43 0.14 0.67 16.10 63.38 20.52
C2 1.9823 0.8081 0.6768 0.00 0.30 0.51 0.81 0.00 36.93 63.07
C2 2.1108 0.8179 0.7142 0.10 0.24 0.58 0.92 10.79 25.71 63.51
C3 3.1521 1.8019 1.5781 0.16 0.51 1.29 1.96 7.98 25.98 66.05
C3 2.8504 1.6117 1.3169 0.04 0.67 0.94 1.66 2.70 40.45 56.85
CC1 2.9839 0.8368 0.7042 0.95 0.30 0.54 1.79 53.25 16.84 29.91
CC1 2.5979 0.7707 0.6566 0.63 0.26 0.51 1.40 45.10 18.47 36.43
CC2 3.821 2.1826 1.7499 0.44 0.98 1.20 2.63 16.92 37.43 45.65
CC2 5.2828 3.274 2.6318 0.81 1.46 1.81 4.09 19.93 35.70 44.38
Mean 0.36 0.57 0.84 1.77 19.20 33.43 47.37
STDEV 0.36 0.40 0.52 1.08
Area of Sample = 32 cm2 Area of Geotextile = 201 cm2 Mvs = Mass of volatile solids Mcalc = Mass of calcite Mrefr = Mass of refractory material
98
For the five laboratory columns , the average mass of calcite precipitate within the
uppermost geotextiles during the life of the experiment can be calculated using
Table 5.2, the area of the geotextile samples (32 cm2), and the area of the whole
geotextile (201 cm2). This average mass of calcite was 3.4g, which in turn
corresponds to 33% of the total dry weight of the material deposited within the
geotextile. Similarly, the mass and percentage of dry weight of volatile solids, and
“refractory material’, can be calculated. They correspond to 2.1g or 19% and 4.8g
or 47% respectively.
Fleming et al.( 1999) also found that the amount of organic matter as a percentage of
clog weight varies from 2 up to 8.5%.Thus between 42 and 58% by weight would be
composed of inorganic material inherent to the refuse. Table 5.3 summarizes the
laboratory findings and the ones found by Fleming et al (1999).
It is important to note that in this study, the mineral deposits that were analyzed were
taken from the geotextile filter overlying and protecting the drainage media. The
higher proportion within the geotextile of non-calcite “refractory” mineral deposit
tends to support conclusions (i.e. Fleming and Rowe, 2004) that a geotextile filter
overlying the drainage media is effective at protecting the underlying drain from
clogging.
99
Table 5.3 Clogging composition in the laboratory experiment vs. clogging
composition in the findings of Fleming et al. (1999)
Volatile
Solids
Calcite
(CaCO3)
Refractory
material
% by dry weight in
lab experiment 19 33 47
% by dry weight in
Fleming et al. 3.8 58 38.2
Although clog composition is a time varying property (VanGulck and Rowe, 2004),
some comparisons can be drawn form the results listed in Table 5.3.
Firstly, the percentage by weight of volatile solids in the uppermost geotextiles in
this laboratory experiment was greater than that found in field excavations by
Fleming et al. (1999). This likely reflects the higher porosity and surface area of the
geotextiles compared to the landfills drains, from where the clog was extracted.
Studies done by Koerner and Koerner (1992) showed that the hydraulic conductivity
of geotextiles (non-woven needle punched) will decrease by 65 % after being
permeated with leachate with different strength from different landfills in the United
States over 6 months. Koerner (1995) concluded that geotextiles used in landfills
that produce leachate with higher values than 2,500 mg/L of BOD5 should follow
100
field and laboratory analysis prior placement. Reinhart and Chopra (2000) also
recognized the likelihood of geotextile clogging due to bacterial growth.
Geotextiles present an interesting design conflict. Due to their intrinsic
characteristics, they could be useful for leachate treatment working as fixed biofilm
reactors (i.e. high surface area, high porosity), but at the same time bacterial growth
and mineral precipitation within the fabric could hinder leachate flow.
Secondly, the calcite fraction reported by Fleming et al. (1999) is almost double that
found in the geotextile in this laboratory experiment. It is hypothesized that the
external aerobic nitrification reactor may play a role in this phenomenon. The pH of
the aerated effluent, as it was seen in chapter 4, is higher than that of the anaerobic
column discharge. This higher pH would lead to a greater calcite precipitation since
the solubility of calcite is pH dependant. Therefore, calcite may precipitate in a
substantial amount in the aerated reactor.
Thirdly, what is called refractory material (inorganic material other than calcite)
comprises a larger percentage of the total deposited material in the laboratory
experiment than in the exhumations done by Fleming et al. (1999). Reinhart and
Chopra, 2000, concluded that due to fines and particles coming from the waste, a
layer of sand should be placed on top of the geotextile (acting as a filter) and
geotextiles should be used to separate that filter media from the drainage media.
However, Fleming and Rowe (2004) see the amount of non-calcite mineral deposits
101
as a proof of geotextiles effectiveness at protecting the underlying drains from
clogging.
5.2.3 Calculating the Time for the Geotextiles to Clog
Rittman et al (1996), (2003), Fleming et al (1999); Fleming and Rowe (2004); Rowe
et al., (2002) and VanGulck et al. (2003) have shown a direct relationship between
organic matter stabilization, measured as chemical oxygen demand depletion, and
the amount of calcite precipitate. This relationship, termed the “Yield
Coefficient”(Yc), was found to range from 0.17 to 0.2 mg of calcite precipitate per
mg of COD depleted.
During the present laboratory experiment, the yield coefficient was found to be 0.16
mg of calcite precipitate per mg of COD depleted. The calcite concentration
decreased from for 2,900 mg/L to 150 mg/L as CaCO3 (Ruiz et al. 2004). For a flow
rate of 0.1 L/day the correspondingly rate of calcite precipitation may be taken to be
0.275 g/day per column.
In order to determine how long it will take for a geotextile to clog under the rates of
calcite precipitation experienced during the laboratory testing, the following
calculations were performed.
102
The density of calcite at 25 ˚C is 2.7 g/cm3. The experiment showed a calcite
production per column of 0.275 g/day. Thus, the total volume of calcite deposited in
each column per unit area per day (Vc) can be estimated in the next fashion:
Vc (Volume of calcite deposited in each column / day*cm2) = (0.275 g/day)/ (2.7
g/cm3*182.4 cm
2) = 5.6*10
-4 cm
3/ cm
2*day
The analysis of the clog material deposited within the uppermost geotextiles
(presented in section 5.2.2) shows that the mass of calcite precipitated over 100 days
was 3.3 g or 0.033 g/day. This allows the quantification of the total amount of calcite
deposited within the uppermost geotextiles per unit area per day:
Vcg (Total calcite deposited in geotextiles/ day*cm2) = (0.0331g/day) /(2.7g/ cm
3
*182.4 cm2)= 6.7*10
-5 cm
3/ cm
2*day. Thus, it may be seen that (6.7/54) = 12% of
the calcite was precipitated in the uppermost geotextiles.
Further, the initial porosity and thickness of the geotextile are 90 % and 0.3 cm,
respectively. Thus, the volume of the geotextile available to be filled per area
Vgeotextile (Volume of geotextile available to be filled/cm2) = 0.3 cm*0.9 = 0.27
cm3/cm
2
103
With all this information, and based on the amount of calcite deposited in the
geotextiles, the time for the void space in the uppermost geotextile to be completely
filled by calcite would be:
tg =( 0.27 cm3/ cm
2 )/( 6.7*10
-5 cm
3/ cm
2*day) = 4,030 days = 11 years (approx.)
This result shows that it would take up to 11 years for an underlying geotextile to
clog according to the amount of calcite deposited within the uppermost geotextiles
used in the laboratory experiment. The clogging time would be dramatically
shortened if all the calcite produced by organic matter stabilization were to be
trapped in the geotextile fabric.
It is important to notice that the previous “time to clog” calculation accounts for
neither biological nor “refractory” clogging. The calculations were done in this way
to reflect the fact that the biofilm should reach a near steady state, and as the overall
pore volume decreases, there will be a time in the LCS when the leachate mounding
above the geotextile will create a sufficiently high advective flow and leachate will
be able to flow through this biological mass by shearing it. Similarly, for the
refractory material, it is hypothesized that since such material largely consists of
fines from overlying refuse etc, there will be an initial supply of refractory mineral
particles and the supply of such material would decrease over time. Further, it is
difficult (and beyond the scope of the experiment) to develop a model that accounts
for the fines and the solid material coming out of the refuse.
104
Clearly, there is a connection between geotextile clogging and the amount of organic
matter stabilization. However, other factors could influence the location of mineral
precipitation. In the experiment for instance, the amount of calcite precipitate within
the geotextiles was only on average 12 % of the total amount of calcite precipitate.
The balance of the mineral was deposited in the refuse and/ or the gravel-sized
porous media and/or in the aerated nitrification tank.
Calcite precipitation and its accumulation can occur at different places since they not
only depend on the amount of calcium-carbonate available to precipitate (solubility),
but also, rely strongly on the carbonate equilibrium with the surrounding
environment and the pH conditions (Jefferis and Bath 1999; Fleming et al. 1999;
Fleming and Rowe 2004; Rittman et al.1996; 2003). These factors may vary over
time and not be the same all everywhere in the system. For instance, in chapter 4, it
was shown that pH was substantially higher in the aerated effluent than that of the
column discharge. This situation could lead to higher levels of calcite precipitation
in the aerated nitrification reactor. Furthermore, a recirculation mode of operation
could also play a role in buffering the fresh leachate that is being produced by the
ongoing refuse degradation (or as it was the case in the experiment, buffering the
synthetic leachate being input at the top of the columns). All these mechanisms were
involved in the laboratory experiment, and, it is hypothesised, they explain the
difference between the amount of calcite deposited within the geotextiles and the
total amount of calcite precipitated.
105
5.2.4 Consequences of Geotextile Clogging
Although a decrease of more than one order of magnitude (as experienced by the
uppermost geotextiles in the laboratory columns) seems to be a considerable drop in
hydraulic conductivity, the critical issue, is how well the LCS can work under such
circumstances (lower geotextile hydraulic conductivity). A few calculations are
presented below in regard to landfill serviceability.
Based on Darcy’s equation, geotextile hydraulic conductivity can be determined by:
K = -(qo*∆z) /∆h (5.1)
Using Darcy’s equation for a typical landfill infiltration rate (qo) of 0.15 m3/m
2/yr, a
geotextile thickness (∆z) of 0.3 cm, and geotextile hydraulic conductivity (K) of
0.0021cm/sec, leachate mounding, and (∆h) will account for 6.8*10-5 cm. This value
is much smaller than the standard industry requirement of a maximum of 30 cm for
leachate build-up above a landfill LCS (Subtitle D, USA EPA). In order to have a
build-up of head of concern (e.g. 150 cm), for a 0.3 cm thick geotextile, its
hydraulic conductivity would have to be lower than 10-9 cm/sec which would require
a decrease of six orders of magnitude.
Although, the hydraulic conductivity results obtained in the laboratory experiment
admittedly are not enough in order to forecast the leachate mounding in a real
106
landfill, they do highlight the issue of a likely decrease in the permeability of the
geotextiles overtime, and the possibility that such a decreased permeability can be
controlled and ultimately beneficial for the purpose of leachate treatment.
5.2.5 Effects of Leachate Treatment on the Hydraulic Performance of the
Drains.
Figure 5.5 shows a view of one section of the column C2. The autopsy of the
columns showed blackened spots, most likely reflecting the path that the leachate
Figure 5.5 Black slime on drains
Bioslime
107
took while going from top to bottom of the columns. However, most of the
biological activity was developed in the uppermost geotextiles due to its higher
surface area and porosity.
Figure 5.6 depicts the gravel material shown in figure 5.5 after being hardened with
a non calcium based epoxy. The objective of this was to be able to take cross-
sections (such as the one shown in Figure 5.7) and examine the pore spaces, as well
as to enable microscopic observations and mineral analyses of the precipitate coating
the granular drainage media.
Figure 5.6 Hardened gravel material
108
Figure 5.7 Cross sectional view of the gravel material
The overlying geotextiles acted as a filter and promoter of bacterial growth,
generating most of the leachate stabilization as well as trapping most of the by-
products, leaving the hydraulic properties of the drains intact.
The implications of these findings are important for the operation bioreactor
landfills. Leachate stabilization is linked to bacterial growth and mineral
precipitation. Both factors, as it was seen before, can affect the permeability of the
LCS and especially of the geotextiles. The potential for significantly decreased
permeability of the geotextiles filters has to be addressed in order for a landfill to
take advantage of the benefits of working as a bioreactor.
109
5.3 Potential of Recycled Concrete as a Drainage Material in a Landfill.
A major part of any LCS is the granular drainage medium. Natural gravel deposits
are the primary source of such drainage material in North America. However, in the
interest of preserving these deposits and in recycling used material, the use of
crushed concrete as a drainage medium was investigated.
The purpose of using two different drainage mediums in this laboratory experiment,
was to compare the performance differences (if any) between natural stone and
crushed concrete. The crushed concrete had a somewhat smaller nominal diameter
than the natural stone used in the top drainage layer (See chapter 3, section 3.2).
5.3.1 Fines Accumulation
Before the crushed concrete could be used it was necessary to screen the fines from
the concrete. The screened concrete particles were then placed in the columns and
the experiment began. It was immediately evident that a significant quantity of
concrete fines had migrated in and partially clogged the geotextile immediately
below the crushed concrete, as shown in Figure 5.8. It was then necessary to wash
the concrete prior to use as drainage media to ensure sufficient removal of fines.
Clear stone, on the other hand, did not exhibit such problems.
110
Figure 5.8 Geotextile clogged with fines originating from sieved, but not washed,
crushed concrete.
This finding has two implications when using crushed concrete as a drainage
medium. The first one is that crushed concrete may need costly preparation before it
can be used in the LCS, and the second one is that fines coming into the drainage
system from the crushed concrete, may provide nucleation sites for the precipitation
of calcite (CaCO3) which would be expected to contribute to more severe clogging
of the LCS (Fleming et al, 1999).
These fines could accumulate and cause precipitation in three locations: the drainage
medium (the crushed concrete), adjacent to or near the drainage pipes, or within a
geotextile filter placed to separate two layers of drainage material.
111
5.3.2 Leachate Mounding
The hydraulic head above the first geotextile (directly beneath the refuse) was
monitored within all the columns throughout the entire experiment duration. It was
found that these head measurements were consistently higher in the two columns
filled with crushed concrete, relative to the head measured in the columns packed
with natural gravel (Figure 5.9). While the precise cause is not known, it seems
likely that the crushed concrete, through a physical or chemical process is associated
with more severe clogging of the overlying geotextile.
0
5
10
15
20
25
30
35
0 20 40 60 80 100 120
Elapsed time (days)
Leachate Mounding Height (cm)
C1 CC1 C2
CC2 C3
Figure 5.9 Graph representing leachate mounding above the top geotextile
112
5.3.3 Alteration of Leachate pH
Throughout the experiment, the discharge pH in the columns filled with crushed
concrete was higher than the columns which used natural stone. This is likely due
to interaction between the naturally alkaline concrete and the leachate. This effect is
a concern as the solubility of calcite is lowered as pH is raised. The resulting
consequence is potentially greater precipitation of calcite. Such calcite precipitation
would further contribute to the clogging of the LCS, since LCS clog material has
been shown to be composed largely of calcite (Fleming et al, 1999, Bennet et al,
2000, Fleming and Rowe, 2004). Figure 5.10 illustrates the history of the leachate
pH for each column.
8
8.5
9
9.5
0 10 20 30 40 50 60 70 80 90 100
Elapsed time (days)
pH
pH C1 pH CC1
pH C2 pH CC2
pH C3
Figure 5.10 pH history of leachate. (See chapter 4: pH in anaerobic column
discharge)
113
5.3.4 Durability of Aggregate
When the experiment was completed, one stone column and one crushed concrete
column were impregnated with epoxy and upon hardening, cut into slices for
examination of the cross-sections.
Inspection of the sections revealed a band of black discolouration permeating into
both the crushed concrete and the clear stone. Figure 5.11 and 5.12 display
photographs of weathered stone and weathered crushed concrete respectively. This
discolouration penetrated more deeply into the crushed concrete and limestones, and
to a lesser degree, sandstones and igneous rocks.
Figure 5.11 Slice of weathered stone with rock types labelled
Limestone
Granite
Sandstone
Black weathering
114
Figure 5.12 Slice of weathered crushed concrete.
The discoloured concrete exhibited signs of reduced durability. Upon visual
inspection of the discoloured regions, the concrete was crumbly and small grains
could be dislodged by running a finger across the surface. Though the natural stone
also had black discolouration, it did not exhibit any obvious signs of weakening as a
result of such alteration. The discoloration in the concrete can also be seen by
comparison of the photomicrographs of non-weathered and weathered concrete
represented by Figures 5.13 and 5.14 respectively.
Black weathering
115
Figure 5.13 non-weathered concrete (100 µm)
Figure 5.14 weathered concrete(100 µm)
116
In order to quantify the seeming crumbly appearance of the weathered concrete, a
slake test was carried out using the non-weathered and weathered concrete. Table
5.4 shows the results of the test.
Table 5.4 Slake test for non-weathered and weathered concrete
Type of Concrete Weight before
slake test (g)
Weight after slake
test (g)
Variation (%)
Non-weathered 498.46 493.3 1
Non-weathered 464.20 459.81 0.9
CC1 weathered 471.17 469.16 0.4
CC1 weathered 483.46 480.12 0.7
CC2 weathered 480.95 477.06 0.8
CC2 weathered 465.8 462.92 0.6
No significant variation was observed among the different concretes. The
explanation for these results may lie in the low speed at which the test is performed.
The visual weathered-crumbly appearance was not reflected in a test that is designed
for softer materials in order to show a sizable difference in weight.
5.5 Conclusions
After 100 days of percolation by high-strength synthetic leachate, the uppermost
geotextiles experienced a decrease in hydraulic conductivity of around 1.2 orders of
magnitude as an average for the laboratory columns. The reason for this
permeability reduction was clogging by overlying wet refuse, calcite precipitation
and biological growth.
117
Based on the amount of refractory material found within the geotextile, it is likely
useful to separate the geotextile from the refuse. A layer of sand in between can be a
viable option in order to avoid having particles from the waste getting trapped in the
geotextile fabric.
Based on extrapolating the data from the laboratory columns, it is hypothesized that
an increased rate of recirculation in a landfill will more quickly decrease the
hydraulic conductivity of the geotextile, thus potentially increasing leachate
mounding over the geotextile filter. Leachate recirculation has been proven to
accelerate waste stabilization which in turn is closely related to calcite precipitation
and bacterial growth (therefore to geotextile permeability). This presents a potential
design and operational challenge for landfills operated as bioreactors.
Even though a faster waste decomposition means a greater amount of carbonates
released into leachate, calcite deposition also depends upon other factors such as pH,
CO2 partial pressure, temperature and calcium availability. Since all these
components are not always constant, the place and intensity of such deposition can
vary widely throughout the landfill.
The crushed concrete contained a substantial content of fines, even after screening to
a uniform size of approximately 30 mm. Unless these fines are removed, the greater
fines content would potentially result in accelerated clogging within filters, within
118
the drainage media itself, or within the perforated leachate collection pipes, which
would increase the cost of preparation of the material for use.
The higher leachate pH within the crushed concrete drainage media may contribute
to more severe clogging of the LCS, due to the fact that with increasing pH, there is
a decrease in the solubility of calcium carbonate (the main mineral component of
clog material).
The columns containing the crushed concrete exhibited greater mounding of leachate
above the primary geotextile than the natural stone counterparts. This may have
been associated with either or both of the above-mentioned factors.
When considering the lifespan of the landfill, the questionable durability of the
concrete may be a serious limitation. After only approximately one hundred days of
operation, the crushed concrete appeared to have undergone noticeable weathering
when compared to natural stone. The performance of the crushed concrete over a
lifetime of 50 to 100 years or greater cannot be assured. Further study is required
with respect to this phenomenon, however until this issue has been resolved, it is
recommended that substantial caution be exercised in using crushed concrete as an
alternative granular drainage media in municipal landfill leachate collection systems.
119
Chapter 6 Discussion and Recommendations
6.1 Introduction
This chapter discusses and summarizes the important conclusions reached in
previous chapters. It also provides recommendations that will guide future
investigations in the field of solid waste management.
6.2 Discussion
The management and operation of Municipal Solid Waste (MSW) landfills has
evolved greatly over the past 20 years. Landfill settlements, landfill gas (LFG)
management, and ground water contamination are some of the challenges facing
landfill operators and/or owners in meeting increasingly stringent environmental
regulations. Landfills have to be monitored until their outputs have stabilized (i.e.
leachate, landfill gas, settlements), and no longer represents a threat to the
environment. If solid waste is left to degrade without any intervention, the time span
to reach stabilization would be much longer than the landfill can be used for waste
collection.
120
Thus, every effort aimed at shortening the time necessary to reach stabilization is
worthy. As was explained in chapter two, the benefits for landfill owners and
operators are numerous.
The bioreactor landfill is a novel approach design to reach rapid stabilization of the
solid waste. The main feature of this system is leachate recirculation; however,
leachate recirculation also brings challenges to landfills such as potential slope
instability, increased ammonia loads in the leachate, and the possibility of clogging
of the Leachate Collection System (LCS). Despite these challenges, the central
argument of this thesis is that the LCS can play a fundamental role in achieving
rapid solid waste stabilization.
Several researchers have conducted field and laboratory studies on the clogging of
LCS under a number of conditions. The main conclusion of these studies is that
clogging is a process that occurs naturally, as a by-product of organic matter
degradation. Based on this conclusion and on the works of Fleming et al. (1999) and
Fleming et al. (2002), this thesis argues that a LCS can act as a fixed biofilm reactor.
A landfill LCS possesses surface area and porosity that makes it suitable for being
colonized by microorganisms, which in turn will decompose contaminants (nutrients
for them), converting complex potentially harmful compounds into more simple,
valuable and/or harmless forms, such as carbon dioxide, nitrogen gas, and methane
gas.
121
A fine equilibrium is the key for the workability of this proposed argument.
Biological clogging means that leachate is being treated but also means that the free
drain, and therefore subsequent collection of the leachate, can be hampered. Thus, a
LCS design that promotes bacterial settlement and allows free flow of the leachate
will meet this fine equilibrium.
In this regard, five PVC columns were packed with porous media (granular material
and geotextile), resembling those found in a landfill LCS. The size and location of
these porous media, as well as the operating parameters such as leachate
recirculation rates were intended to be similar to those in Municipal Solid Waste
(MSW) leachate collection and treatment. An external aerated reactor connected to
each column, was also set up with the primary function of achieving nitrification of
the ammonia present in the leachate, an intermediate step in the conversion of this
ammonia to harmless nitrogen gas.
The specific objectives of the laboratory experiment were to evaluate the removal of;
Chemical Oxygen Demand (COD), nitrogen in the form of ammonia, and some
inorganic material (Calcium), as well as to evaluate the hydraulic performance of the
modified drainage LCS after the physical, chemical, and biological process,
associated with that removal, took place.
The results regarding leachate quality were very satisfactory. Approximately 97% of
the initial COD was removed. 98% of the ammonia was converted to nitrogen gas
122
and 95% of the original calcium concentration was precipitated in the form of
calcium carbonate. The hydraulic properties of the upper most geotextiles were
affected by bio-slime formation along with fine material and mineral precipitate
(mainly calcite) that were entrapped within the geotextiles. As a result, the
uppermost geotextiles experienced a decrease in hydraulic conductivity of
approximately 1.2 orders of magnitude on average. No clogging was observed in
the natural gravel, in the crushed concrete, or in the other geotextiles. There are two
reasons for this. First, the surface area of the natural gravel and the crushed concrete
is smaller than that of the geotextiles, making it more difficult for the bacteria to
attach and grow. Second, leachate strength is the highest when it is passing through
the uppermost geotextile, thus most of the bacterial activity is developed at that
location.
The use of concrete as a drain material for a landfill is not recommended. Noticeable
weathering appearance was observed on the material, as well as a higher pH of the
column discharge when compared to that of the columns containing natural gravel.
When considering the lifespan of the landfill, the questionable durability of the
concrete may be a serious limitation.
The biogas produced during the laboratory experiment was in constant amounts and
its composition presented positive implications from the perspective of landfill gas
management and operation. Firstly, the levels of methane gas were large enough to
be considered as a possible source of energy. Secondly, the level of carbon dioxide
123
was very low compared to that found in typical landfills. Carbon dioxide is a
corrosive agent. Its low concentration might extend the life of the landfill gas (LFG)
collection system.
6.3 Recommendations
Future researchers who continue with the line of experimentation described in this
thesis should find the following recommendations useful:
1. A separator between the filter and the refuse should be incorporated in the
design of landfill LCS. In the laboratory experiment, almost 50 % of the material
entrapped in the geotextile fabric is thought to come from the waste. The type of
material used for the separator will be determined based largely on the operating
conditions of the landfill (i.e. rates of recirculation).
2. The use of an external aerated nitrification reactor may be useful not only for
nitrogen treatment but also for decreasing the amount of carbonate present in
leachate and therefore the amount of calcite precipitation. The aeration of the tank,
coupled with the unbalanced carbonate system (supersaturated with carbonates) may
decrease the amount of carbonates present in the reactor by releasing it to the
atmosphere.
3. Given the rather substantial cost savings that might be achieved through the
use of crushed concrete in lieu of natural gravel, a limited extension of this study
124
should be conducted in order to determine to what extent crushed concrete should be
used in landfill drainage only with extreme caution or not at all.
4. The overall results of the laboratory experiment were positive, especially for
nitrogen and COD removal rates. It is recommended, that the knowledge acquired
during this laboratory experiment be applied to a pilot experiment under real landfill
conditions.
125
References
Armstrong, M.D. 1998. Laboratory program to study clogging in a leachate
collection systems. ME.Sc thesis, The University of Western Ontario, London, Ont.
Azevedo, B.D., Mavinic, D.S., and Robinson, H.D. 1995. The effect of ammonia
loading and operating temperature on nitrification and denitrification of a high
ammonia landfill leachate. Canadian Journal of Civil Engineering.22 : 524-534.
Barlaz, M.A., Schaefar, D.M., and Ham, R.K. 1989. Bacterial Population
Development and Chemical Characteristics of Refuse Decomposition in a Simulated
Sanitary Landfill. Appl.Env. Microbiol. 55(1): 55-56.
Barlaz, M.A., Rooker, A.P., Kjeldsen, P., Gabr, M.A., and Borden, R.C. 2002.
Critical evaluation of factors required to determine the post-closure monitoring
period at solid waste landfills. Environmental Science and Technology. 36 (16):
3457-3464.
Bennet, P.J., Longstaffe, F.J., and Rowe R.K., 2000. The stability of dolomite in
landfill leachate collection systems. Canadian Geotechnical Journal. 37: 371-378.
Brune, M., Ramke, H.G., Collins, H., and Hanert, H.H. 1991. Incrustation processes
in drainage systems of sanitary landfills. Proc. Sardinia 91, 3rd International
Symposium on Sanitary Landfills, CISA, Cagliary, Italy, pp.99-1035
Burton, S.A.Q., and Watson-Craick, I.A. 1998. Ammonia and nitrogen fluxes in
landfill sites: applicability to sustainable landfilling. Waste Management and
Research. 16 (1): 41-53.
Cooke, A.J., Rowe, R. K., Rittmann, B. E., VanGulck, J., and Millward S.C. 2001.
Biofilm growth and mineral precipitation in synthetic leachate columns. ASCE
Journal of Geotechnical and Geoenvironmental Engineering. 127(10):849-856.
Christensen, T.H., and Kjeldsen, P. 1989. Basic Biochemical processes in landfills.
In: Sanitary Landfilling: Process, Technology, and Environmental Impact, ed. T.H.
Christensen, R. Cossu and R. Stegmann. Elsevier Applied Science, London, UK.
Christensen, T.H., Kjeldsen, P. and Stegmann, R. 1992. Effects of landfill
management procedures on landfill stabilization and leachate and gas quality. Ed
T.H. Christensen, R. Cossu and R. Stegmann. Academic Press, London, UK.
126
Farquhar, G.J. Leachate: Production and Characterization. 1989. Canadian
Geotechnical of Civil Engineering. 16: 317-325.
Fleming, I.R., Rowe, R.K.,and Cullimore, D.R. 1999. Field observations of clogging
in a landfill leachate collection system. Canadian Geotechnical Journal.36(4):685-
707.
Fleming, I.R., and R.K. Rowe. 2000. Toward Sustainable Leachate Treatment in
Landfills. CSCE/CGS Specialty Environmental Conference, London, Ont., June,
p15-20.
Fleming, I.R., Rowe, R.K., Dewacle, P., and Benda, E. 2002. Evolution of leachate
strength in municipal landfills; can leachate be treated in-situ? In: Proceedings form
the Solid Waste Association of North America. 7th Annual landfill symposium.
Louisville, Kentucky.
Fleming, I.R., and Rowe, R.K. 2004. Laboratory studies of clogging of landfill
leachate collection and drainage systems. Canadian Geotechnical Journal. 41(1):134-
153.
Gallagher, E.M. 1998. Biological clogging of geocomposites exposed to raw landfill
leachate over an extended period. Waste Management Research 16(5):421-429.
Halse, Y., Koerner, R.M., and Lord, A.E. 1987. Effect of High Levels Alkalinity on
Geotextiles. Part 1: Ca (OH) Solutions. Geotextiles and Geomembranes. 5:261-282.
Halse, Y., Koerner, R.M., Lord, A.E., 1987. Effect of High Alkalinity Levels on
Geotextiles. Part 2: NaOH Solution. Geotextiles and Geomembranes. (6):295-305.
Harris, J.M., Purschwitz, and D.E., Goldsmith, C.D. 2000. Leachate treatment
options for sanitary landfills. Intercontinental Landfill Research Symposium. Lulea,
Sweden.
Jetten, S.M., Wagner, M., Fuerst, J., Van Loosdrecht, M., Kuenen, and G., Strous,
M. 2001. Microbiology and application of the anaerobic ammonium oxidation
(‘anammox’) process. Current opinion in biotechnology. 12: 283-288.
Jefferis, S.A., and A. Bath., 1999. Rationalising the debate on calcium carbonate
clogging and dissolution in landfill drainage materials. Proceedings Sardinia 99,
Seventh International Waste Management and Landfill Symposium, S. Margherita di
Pula, Cagliari, Italy.
Kiely, G. 1997. Environmental Engineering. McGraw-Hill.
127
Knox, K. 1985. Leachate treatment with nitrification of ammonia. Water Research.
19:895-904.
Koerner, R.M., Lord, A.E., and Halse, Y. 1988. Long-term durability and aging of
geotextiles. Geotextiles and Geomembranes. 7:147-158.
Koerner, G.R., and Koerner, R.M. 1992. Leachate flow rate behaviour through
geotextile and soil filters and possible remediation methods. Geotextiles and
Geomembranes. 11:401-430.
Koerner, R.M. 1994. Performance evaluation of geotextiles filters used in leachate
collection of solid waste landfills. PhD dissertation, Drexel University, Philadelphia,
PA.
Koerner, G.R. and Koerner, R.M. 1995. Leachate clogging assessment of geotextile
(and soil) landfill filters. US EPA Report, CR-819371.
Leckie, J.O., Pacey, J.G. and Halvadakis, C. 1979. Landfill management with
moisture control. ASCE, Journal of Environmental Engineering Division.105: 337-
355.
Leushner, A.P. 1989. Enhancement of degradation: laboratory scale experiment. In
Sanitary Landfilling: Process, Technology, and Environmental Impact, ed. T.H.
Christensen, R. Cossu and R. Stegmann. Academic Press, London, UK.
Lin, Yin-Feng, and Chen, Kuo-Cheng. 1995. Denitrification and methanogenesis in
a co- immobilized mixed culture system. Water Research 29(1):35-43.
McBean, E.A., F.A. Rovers, and G.J. Farquhar. 1995. Solid Waste Landfill
Engineering and Design. Prentice Hall. Englewood Cliffs, N.J.
Mlynarek, J., and Vermeersch, O. Filter criteria for geotextile materials in soil
filtration and leachate collection systems. 1999. Proceedings Sardinia 99, Seventh
International Waste Management and Landfill Symposium, Cagliari, Italy, vol. 3,
pp.233-240.
Onay, T. and Pholand, F. 1998. In Situ Nitrogen Management in Controlled
Bioreactor Landfills. Water Research. 32 (5):1383-1392.
Paksy, A., Powrie, W., Robinson, J.P., and Peeling. 1998. A Laboratory
investigation of anaerobic microbial clogging in granular landfill drainage media. L.
Geotechnique 48(3):389-401.
Palmisano, A.C. Barlaz, M. 1996. Microbiology of solid waste. CRC press. New
York, N.Y.
128
Parkin G.F., and Owen, W.F. 1986. Fundamental of anaerobic digestion of waste
water sludges. ASCE Journal of Environmental Engineering. 112 (5):867-920.
Peeling, L., Paksy, A., Robinson, J.P. and Powrie, W. 1999. Removal of volatile
acids from synthetic landfill leachate by anaerobic biofilms on drainage aggregates:
A laboratory study. Waste Management and Research. 17(2):141-149.
Pohland, F.G. 1980. Leachate recycle as a management option. J.Environ.Eng.Div.
ASCE 106(6): 1057-1069.
Pohland, F.G. and Yousfi, B.A. 1994. Design and operation of landfills for optimum
stabilization and biogas production. Water Science and Technology. 30 (12):117-
124.
Pohland F.G. 1995. Landfill bioreactors; historical perspective, fundamental
principles, and new horizons in design and operation. In Landfill Bioreactor Design
and Operation Sem. Proc., EPA/600/R-95/146,pp.9-24.
Pohland F.G., and Kim, J.C. 1999. In situ anaerobic treatment of leachate in landfill
bioreactors. Water Science and Technology. 40 (8):203-210.
Polarczyk, K.W. 2000. Geosynthetics and Geosystems in Hydraulic and Coastal
Engineering. A:A. Barkena/Rotterdam/Brookfield.
Price, G.A., Barlaz, M.A., and Hater, G.R. 2003. Nitrogen management in bioreactor
landfills. Waste Management. 23: 675-688.
Quian, X., Koerner, R.M., and Gray, D.H. 2002. Geotechnical aspects of landfill
design and construction. Prentice Hall.Upper Saddle River, N.J.
Ramke, H.G. 1989. Leachate Collection Systems, Published in Sanitary Landfilling:
Process, Technology and Environmental Impact, Academic Press.
Reddi, L.N. 1997. Particle transport in Soils: Review of Significant Processes in
Infrastructure Systems. Journal of Infrastructure Systems 3(2): 78-75.
Reinhart, D.R, and Chopra, M.B. 2000. MSW Landfill Leachate Collection Systems
for the New Millennium. State University System of Florida. Florida Center for
Solid and Hazardous Waste Management. University of Florida. Report #00-13
Reinhart, D.R., McCreanor, P.T. and Towsend, T. 2002. The bioreactor landfill: Its
status and future. Waste Management and Research. 20 (2):172-186.
Rittmann B. E, Fleming, I.R. and Rowe, K.R. 1996. Leachate chemistry: its
implication for clogging. North American Water and Environmental Congress ’96 ,
129
Anaheim, CA. paper 4(CD ROM), session GW-1, Biological processes in ground
water quality.
Rittmann B. E., Banaszak J.E., Cooke A., and Rowe, R. K. 2003. Biogeochemical
evaluation of mechanisms controlling CaCO3. precipitation in landfill leachate-
collection systems. ASCE Journal of Environmental Engineering.129 (8): 723-730.
Rollin, A.L., Lombard, G. 1988. Mechanisms affecting long-term filtration
behaviour of geotextiles. Geotextiles and Geomembranes. 7:119-145.
Rowe, R.K., I.R. Fleming, M.D. Armstrong D.R. Cullimore, B.E. Rittmann, P. Bennett,
and F.J. Longstaff, 1997b. Recent Advances in Understanding the Clogging of
Leachate Collection Systems. Proceedings Sardinia ‘97, Sixth International Landfill
Symposium, Cagliari, Italy, pp. 383-390.
Rowe, R.K., and I.R. Fleming. 1998. Estimating the Time for Clogging of Leachate
Collection Systems. Proc. 3rd International Congress on Environmental Geotechnics.
Lisbon, Portugal. pp 23-28.
Rowe, R.K, Armstrong, M.D, and Cullimor C.D. 2000. Particle size and clogging of
granular media permeated with leachate. Journal of Geotechnical and
Geoenvironmental Engineering. 126( 9):775-786.
Rowe, R.K., VanGulck, J.F., and Millward, S.C. 2002. Biologically induced
clogging of a granular medium permeated with synthetic leachate. Journal of
Environmental Engineering and Science. 1:135-156.
Rowe, R.K., Armstrong, M.D., and Cullimore, D.R. (2000a). Particle Size and Clogging
of granular Media Permeated with Leachate. ASCE Journal of Geotechnical and
Geoenvironmental Engineering. 126(9):775-786.
Rowe, R.K., Armstrong, M.D. and Cullimore, D.R. (2000b). Mass Loading and the Rate
of Clogging due to Municipal Solid Waste Leachate, Canadian Geotechnical Journal.
37(2):355-370.
Ruiz, E.F, Fleming, I.R., Putz, G. Passive treatment of municipal landfill leachate in
a granular drainage layer. 2005. Submitted to the Journal of Environmental
Engineering and Science. NRC.
Shiskowsky D.M., and Mavinic, D.S. 1998. Biological treatment of a high ammonia
leachate influence of external carbon during initial startup. Water Research.
32(8):2533-2541.
Schmidt, I., Sliekers, O., Scmid, M., Cirpus, I., Strous, M., Bock, E., Kuenen, G.,
and Jetten, M.S.M. 2002. Aerobic and anaerobic ammonia oxidizing bacteria-
competitors or natural partners. FEMS Microbiol.Ecol. 39: 175-181.
130
Stegmann, R., and Spendlin, H.H. 1986. Research activities on enhancement of
biochemical process in sanitary landfills. Waste Management and Research. 1:201-
211.
Stegmann, R., and Spendlin, H.H. 1989. Enhancement of degradation: German
experiences. In: Sanitary Landfilling: Process, Technology, and Environmental
Impact, ed. T.H. Christensen, R. Cossu and R. Stegmann. Academic Press, London,
UK.
Tchobanoglous, G., Theisen H., and Vigil, S. 1993. Integrated solid waste
management: Engineering principles and management issues. McGraw-Hill Inc.
Tiedje, J.M. 1988. Ecology of denitrification and dissimilatory nitrate reduction to
ammonium. In: Zehnder, J.B.A. (Ed), Biology of Anaerobic Microorganism. John
Wiley and Sons. NY, pp. 179-244.
US EPA Process design manual for nitrogen control. 1975. Technology Transfer.
Washington, DC United States.
US EPA Waste water technology fact sheets: Sequencing batch reactors. 1999.
Office Water, Washington, DC United States.
US EPA Municipal solid waste in the United States: 2000 facts and figures. 2002.
Office of Solid Waste and Energy Response, Washington, DC. United States.
VanGulck, J.F., Rowe, R.K., Rittmann B. E, and Cooke, A.J., 2003. Predicting
biogeochemical calcium precipitation in landfill leachate collection systems.
Biodegradation 14(5): 331-346.
VanGulck, J.F. and Rowe, R.K. 2004 Evolution of clog formation with time in columns permeated with synthetic landfill leachate. Journal of Contaminant
Hydrology 75: 115– 139
Warith M.A. and Sharma, R. 1998. Technical review of methods to enhance biological
degradation in sanitary landfills. Water Quality Research Journal. 33 (3): 417-437.